An electrocardiosignal compensation method based on impedance feedback and intelligent clothing

Abstract

The application discloses an ECG signal compensation method based on impedance feedback and intelligent clothing, and belongs to the technical field of intelligent clothing. In view of the defects that the existing textile electrode ECG collection is easily disturbed by sweating and exercise, the compensation mode is single, and the electrode fault tolerance is poor, the application adopts a time division multiplexing architecture, and synchronously realizes ECG collection and skin-electrode interface dual-frequency impedance detection. According to the impedance amplitude, the change rate and the frequency response characteristics, the application divides the electrode contact state into four categories, namely, dry stability, sweating transition, sweat saturation and motion loss, and configures differentiated signal compensation and electrode regulation strategies. The device is equipped with a redundant electrode array and an analog switch matrix, can adaptively optimize the electrodes, dynamically adjusts the parameters in combination with the exercise intensity, and supports sweat saturation prediction and pre-compensation. The application does not need to add detection electrodes, can effectively suppress various ECG noises and signal distortions, has the advantages of collection accuracy, stability and low power consumption, and is suitable for the continuous ECG monitoring demand in complex scenes.

Description

Technical Field

[0001] This invention relates to the field of smart clothing technology, and more specifically to a method for compensating electrocardiogram signals based on impedance feedback and smart clothing. Background Technology

[0002] With the development of smart wearable devices, smart clothing integrating electrocardiogram (ECG) monitoring functions is gradually being applied to scenarios such as sports and health monitoring and remote medical monitoring. Smart clothing typically uses textile electrodes instead of traditional wet electrodes, weaving conductive yarns into the fabric so that they come into contact with the skin when worn to collect ECG signals.

[0003] However, textile electrodes are dry electrodes, and their contact impedance with the skin is easily affected by various factors. During exercise, sweating causes the skin-electrode interface to transition from dry to wet: in the early stages of sweating, sweat begins to wet the stratum corneum, causing a rapid decrease in contact impedance and introducing baseline drift; after heavy sweating, sweat forms a continuous liquid film on the electrode surface, leading to bypass short circuits between adjacent electrodes, resulting in attenuation or even distortion of the ECG signal amplitude. Furthermore, body deformation caused by exercise leads to relative displacement between the electrode and the skin, causing instantaneous fluctuations in contact impedance, manifesting as motion artifacts.

[0004] Existing technologies primarily address these issues from two directions: first, by improving electrode materials and structures, such as using gradient porous composite substrates to achieve waterproofing and sweat wicking, self-cleaning coatings to repel sebum, and biomimetic embroidery structures to increase adhesion; second, by employing fixed-parameter filtering algorithms to suppress noise in the signal post-processing stage. The former is a passive adaptation and cannot respond in real-time to changes in electrode state; the latter, using fixed filtering parameters, suffers from significantly reduced filtering effectiveness when changes in contact impedance alter signal characteristics, and it cannot solve the signal amplitude attenuation problem caused by sweat bypassing.

[0005] Some existing solutions mention adaptive calibration, but they only adjust the gain based on the overall signal-to-noise ratio, without real-time measurement and status identification of the skin-electrode interface impedance. This makes it impossible to distinguish whether signal degradation is caused by sweating, electrode detachment, or motion artifacts, thus hindering the implementation of targeted compensation strategies. Furthermore, current smart clothing typically only has a minimal number of ECG electrodes (e.g., 2-3), lacking redundant alternatives when one electrode makes poor contact. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides an electrocardiogram (ECG) signal compensation method and smart clothing based on impedance feedback. By measuring the impedance of the skin-textile electrode interface in real time and identifying the contact state, the method adaptively adjusts the signal acquisition parameters and switches redundant electrodes, thereby achieving stable acquisition of ECG signals under conditions of exercise and sweating.

[0007] To achieve the above objectives, the present invention specifically adopts the following technical solution:

[0008] An electrocardiogram signal compensation method based on impedance feedback includes the following steps:

[0009] S1. ECG signal acquisition and impedance measurement are performed alternately in a time-division multiplexing manner using a textile electrode array; during the impedance measurement time slot, an AC excitation signal with known amplitude and frequency is injected into at least one pair of textile electrodes, the response voltage at both ends of the electrodes is measured, and the contact impedance value of the skin-electrode interface is calculated.

[0010] S2. Based on the amplitude, rate of change, and frequency response characteristics of the contact impedance, the skin-electrode contact state is classified into one of the following: dry stable state, sweating transition period, sweat saturation state, and motion-induced contact loss state.

[0011] S3. Execute corresponding compensation strategies based on the classification results: When in a dry and stable state, use standard gain and broadband filtering; when in a sweating transition period, reduce the high-pass filter cutoff frequency to suppress baseline drift, and increase the instrumentation amplifier gain to compensate for signal attenuation caused by sweat shunting; when in a sweat-saturated state, switch to redundant electrode pairs and adjust the feedback gain of the right leg drive circuit; when in a motion-induced contact loss state, narrow the filter bandwidth and reduce the sampling rate, and switch to the redundant electrode pairs with the lowest contact impedance.

[0012] Further, in step S1, the AC excitation signal includes a low-frequency excitation signal and a high-frequency excitation signal. The frequency of the low-frequency excitation signal is 8kHz to 12kHz, and the frequency of the high-frequency excitation signal is 80kHz to 120kHz. The resistance component of the contact impedance is calculated based on the low-frequency excitation response, and the capacitance component of the contact impedance is calculated based on the high-frequency excitation response. When the resistance component decreases rapidly while the capacitance component remains basically unchanged, it is determined to be the sweating transition period. When both the resistance component and the capacitance component fluctuate rapidly, it is determined to be the motion loss state.

[0013] Further, in step S2, the classification method specifically includes: preset impedance thresholds Z_high, Z_low, and a rate of change threshold R_th; when the contact impedance amplitude is greater than Z_high and the absolute value of the rate of change is less than R_th, it is classified as a dry stable state; when the contact impedance amplitude is between Z_low and Z_high and the absolute value of the rate of change is greater than R_th, it is classified as a sweating transition period; when the contact impedance amplitude is less than Z_low and the duration exceeds the set duration, it is classified as a sweat saturation state; when the jump in contact impedance amplitude between adjacent measurement cycles exceeds the set jump threshold, it is classified as a motion-induced contact loss state.

[0014] Further, in step S3, the method for switching redundant electrode pairs is as follows: the textile electrode array includes at least three main electrodes and redundant electrodes arranged adjacent to each main electrode, and each electrode is connected to the signal acquisition channel through an analog switch matrix; in each impedance measurement time slot, the contact impedance of all available electrode pairs is measured sequentially to establish an impedance matrix; when the contact impedance of any electrode in the main electrode pair deteriorates to the trigger threshold, the redundant electrode pair with the best contact impedance is selected from the impedance matrix to replace the deteriorated main electrode, and the selection state of the analog switch matrix is ​​updated.

[0015] Further, in step S3, the feedback gain adjustment method of the driving right leg circuit is as follows: calculate the common-mode voltage between the current working electrode pair in real time, and adjust the feedback resistance value of the driving right leg amplifier in a step manner according to the deviation between the common-mode voltage and the target common-mode voltage; when sweat saturation causes the common-mode rejection ratio to decrease, increase the feedback gain to enhance the common-mode rejection capability; when the dry and stable state is reached, decrease the feedback gain to avoid saturation of the driving right leg circuit.

[0016] Furthermore, the method also includes a prediction compensation step: continuously recording the trajectory of contact impedance change over time; when the impedance is detected to enter the sweating transition period and the impedance change rate shows a continuous downward trend, the time to reach the sweat saturation state is predicted by extrapolation based on the impedance change rate; and the redundant electrode pair is actively switched with a preset advance amount before the predicted arrival, so that the switching operation is completed before the signal is severely degraded.

[0017] Further, in step S1, the time ratio of the ECG signal acquisition time slot to the impedance measurement time slot is (8~12):1, the duration of each impedance measurement time slot does not exceed 0.5ms, and the impedance measurement time slot is inserted between two adjacent ECG sampling intervals; within the impedance measurement time slot, the input terminal of the ECG acquisition channel is grounded through an analog switch, and the excitation signal is injected into the electrode through an independent coupling capacitor.

[0018] The present invention also provides a smart garment, comprising a garment body, a textile electrode array, a main control module, and an analog switch matrix;

[0019] The textile electrode array includes at least three main textile electrodes and at least three redundant textile electrodes. The main textile electrodes are respectively located on the garment body at the intersection of the right subclavian, left subclavian, and left anterior axillary lines with the lower edge of the costal arch (approximately the 6th intercostal space). Each redundant textile electrode is located 2cm to 4cm to the side of the corresponding main textile electrode. Each textile electrode is connected to the module interface on the garment body through conductive yarns woven into the fabric.

[0020] The main control module is detachably installed on the garment body and electrically connected to the textile electrode array through the module interface. The main control module includes: a signal acquisition unit for acquiring electrocardiogram signals; an impedance measurement unit for generating AC excitation signals and measuring response voltage to calculate contact impedance; an analog front end, including an instrumentation amplifier, a programmable gain amplifier, a programmable filter, and an analog-to-digital converter; a right leg drive circuit; and a microcontroller for executing the above compensation method.

[0021] The analog switch matrix is ​​positioned between the textile electrode array and the analog front end. Each textile electrode is connected to the positive input terminal, negative input terminal, or right leg drive output terminal of the analog front end via an analog switch. The microcontroller selects the currently working electrode pair and the right leg drive electrode by controlling the on / off state of the analog switch matrix.

[0022] Furthermore, the main control module also includes a Bluetooth communication unit and an accelerometer. The accelerometer is used to collect human motion acceleration. The microcontroller adjusts the impedance measurement frequency according to the motion intensity output by the accelerometer: when the motion intensity is detected to increase, the impedance measurement frequency is increased to speed up the state response speed; when the motion intensity is detected to decrease, the impedance measurement frequency is decreased to reduce power consumption.

[0023] Furthermore, each textile electrode in the textile electrode array has a ring reference electrode at its edge. The ring reference electrode is concentrically arranged with the corresponding textile electrode and is insulated from it. During impedance measurement, an excitation signal is injected into the textile electrode, and the response voltage is detected from the ring reference electrode, so that the current path is concentrated in the skin area directly below the textile electrode, thereby improving the spatial resolution of impedance measurement.

[0024] The beneficial effects of this invention are as follows:

[0025] 1. This invention measures the skin-electrode interface impedance in real time within smart clothing, and then finely classifies the contact state into four states: dry and stable, excessive sweating, sweat saturation, and loss of contact during exercise. Different compensation strategies are adopted for each state. Compared with existing technologies that fix parameters or only adjust the overall gain, this invention can distinguish different causes of signal degradation and perform targeted compensation, significantly improving the acquisition quality of ECG signals under exercise and sweating conditions.

[0026] 2. This invention uses dual-frequency excitation signals to measure the resistive and capacitive components of impedance respectively. By utilizing the characteristic differences that sweat mainly changes the resistive component and motion displacement simultaneously changes the resistive and capacitive components, a reliable distinction between sweating and motion artifacts is achieved, solving the problem that a single impedance parameter cannot distinguish the cause of degradation.

[0027] 3. This invention sets up redundant textile electrodes and an analog switch matrix on the clothing. When the main electrode deteriorates due to sweat saturation or poor contact, it can switch to the redundant electrode pair with the optimal contact state in real time, bypassing the deteriorated electrode from a physical level rather than relying solely on signal processing compensation, thereby improving the reliability of the system during long-term exercise.

[0028] 4. This invention uses time-division multiplexing, which allows ECG acquisition and impedance measurement to share the same set of textile electrodes, eliminating the need for additional impedance measurement electrodes and avoiding increased clothing complexity and impact on wearing comfort; the impedance measurement time slot is extremely short (not exceeding 0.5ms), inserted into the ECG sampling interval, and its impact on the continuity of the ECG signal is negligible.

[0029] 5. This invention predicts the arrival time of sweat saturation based on the impedance change trajectory and proactively switches redundant electrodes in advance, so that the switching operation is completed before the signal is severely degraded, avoiding signal interruption or quality degradation caused by post-compensation. Attached Figure Description

[0030] Figure 1 This is a schematic diagram of the overall structure of the smart clothing of the present invention;

[0031] Figure 2 This is a block diagram of the main control module;

[0032] Figure 3 This is a schematic diagram of the connection of an analog switch matrix;

[0033] Figure 4 This is a schematic diagram of time-division multiplexing time slot allocation;

[0034] Figure 5 This is a block diagram of the impedance measurement unit.

[0035] Figure 6 This is a schematic diagram of a textile electrode with a ring-shaped reference electrode;

[0036] Figure 7 This is a schematic diagram of the module interface structure;

[0037] Figure 8 This is a schematic diagram of the structure of a textile electrode array.

[0038] Reference numerals: 1. Garment body; 2. Textile electrode array; 3. Main control module; 4. Analog switch matrix; 201a. First main electrode; 201b. Second main electrode; 201c. Third main electrode; 202a. First redundant electrode; 202b. Second redundant electrode; 202c. Third redundant electrode; 203a. Ring reference electrode; 204. Conductive yarn; 205. Module interface; 206. Insulating yarn; 301. Microcontroller; 302. Signal acquisition unit; 303. Impedance measurement unit; 304. Analog front end; 305. Right leg drive circuit; 306. Bluetooth communication unit; 307. Accelerometer; 308. Power management unit; 3031. Direct digital frequency synthesizer; 3032. Voltage-controlled current source; 3033. Instrumentation amplifier circuit; 2051. Insulating base fabric; 2052. Conductive central fabric; 2053. Conductive snap button. Detailed Implementation

[0039] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0040] like Figure 1 As shown, the smart clothing of the present invention includes a garment body 1, a textile electrode array 2, a main control module 3, and an analog switch matrix 4. The garment body 1 is made of elastic knitted fabric in a tight-fitting sportswear style to ensure that the textile electrode array 2 remains in close contact with the skin during movement.

[0041] like Figure 8 As shown, the textile electrode array 2 includes three main textile electrodes and three redundant textile electrodes. The first main electrode 201a is positioned on the garment body 1 below the right clavicle, and the first redundant electrode 202a is positioned 3 cm to the side of the first main electrode 201a. The second main electrode 201b is positioned below the left clavicle, and the second redundant electrode 202b is positioned 3 cm to the side of the second main electrode 201b. The third main electrode 201c is positioned at the intersection of the left anterior axillary line and the lower edge of the ribcage, and the third redundant electrode 202c is positioned 3 cm to the side of the third main electrode 201c. The redundant electrodes are made using the same conductive yarn and weaving process as the main electrodes. Each main electrode and redundant electrode has a concentric annular reference electrode 203 at its edge. Each electrode is connected to a module interface 205 located at the back neckline of the garment via conductive yarn 204 woven into the fabric.

[0042] like Figure 7As shown, module interface 205 includes an insulating base fabric 2051, a conductive core fabric 2052, and conductive snaps 2053. The insulating base fabric is fixedly connected to the garment body by sewing or heat-pressing. Six conductive core fabrics are arranged in a matrix on the insulating base fabric, and the insulating base fabric is fixedly connected to each conductive core fabric by sewing or heat-pressing. Six conductive snaps are arranged, one concentrically on each conductive core fabric. Each conductive snap is conductively and tightly connected to its corresponding conductive core fabric, and is fixedly assembled onto the conductive core fabric. Each conductive yarn 204 corresponds to a single conductive core fabric, and the end of each conductive yarn 204 is sewn to its corresponding conductive core fabric. The conductive snaps are fixed to the conductive core fabric by riveting, with the riveted parts of the snaps pressing tightly against the conductive core fabric to achieve a conductive and tight connection. The back of the main control module 3 is equipped with corresponding snap-fit ​​terminals that match each set of conductive snaps. The main control module 3 is pressed onto the corresponding conductive snaps of the module interface 205 through the snap-fit ​​terminals, thereby achieving mechanical locking and fixation between the main control module 3 and the module interface 205, as well as electrical conduction of multiple signals. The main control module 3 can be directly removed from the conductive snaps of the module interface 205. After the main control module 3 is removed, the garment body 1 can be directly washed. The snap-fit ​​mechanism between the conductive snaps and the snap-fit ​​terminals is existing technology and will not be described further.

[0043] like Figure 6 As shown, taking the first main electrode 201a as an example, its edge is provided with a ring reference electrode 203a. The ring reference electrode 203a and the first main electrode 201a are arranged concentrically and isolated from each other by an insulating yarn 206. During impedance measurement, an AC excitation signal is injected into the first main electrode 201a, and the response voltage is detected from the ring reference electrode 203a. This concentrates the excitation current path in the skin area directly below the first main electrode 201a, preventing the current from spreading to adjacent electrode areas and improving the spatial resolution of the impedance measurement.

[0044] like Figure 2 As shown, the main control module 3 includes a microcontroller 301, a signal acquisition unit 302, an impedance measurement unit 303, an analog front end 304, a right leg drive circuit 305, a Bluetooth communication unit 306, an accelerometer 307, and a power management unit 308. The main control module 3 is mechanically and electrically connected to the garment body 1 via conductive snaps on the module interface 205. After disassembly, the garment can be washed.

[0045] The signal acquisition unit 302 includes an instrumentation amplifier and a right leg drive feedback circuit for differential amplification of the weak ECG signal acquired by the textile electrodes. The analog front-end 304 includes a programmable gain amplifier, a programmable filter, and an analog-to-digital converter. The programmable gain amplifier has a gain range of 1 to 128 times and can be configured by the microcontroller 301 via a serial interface. The programmable filter is a high-order switched-capacitor filter with a high-pass cutoff frequency configurable from 0.05Hz to 0.5Hz and a low-pass cutoff frequency configurable from 40Hz to 150Hz. The analog-to-digital converter has a sampling rate configurable from 250Hz to 1000Hz and an adjustable resolution from 16 bits to 24 bits.

[0046] like Figure 5 As shown, the impedance measurement unit 303 includes a direct digital frequency synthesizer 3031, a voltage-controlled current source 3032, and an instrumentation amplifier circuit 3033. Under the control of the microcontroller 301, the direct digital frequency synthesizer 3031 generates a low-frequency excitation signal (10kHz) and a high-frequency excitation signal (100kHz). The voltage-controlled current source 3032 converts the excitation signal into a constant current AC signal with an amplitude of 10μA, which is injected into the main textile electrode or redundant textile electrode in the textile electrode array through a coupling capacitor. The instrumentation amplifier circuit 3033 measures the amplitude and phase of the response voltage across the electrode at the excitation frequency and sends this data to the microcontroller 301 to calculate the complex impedance.

[0047] like Figure 3 As shown, analog switch matrix 4 is a 6×4 crossbar switch array. The six channels correspond to three main electrodes and three redundant electrodes, respectively. The four outputs are connected to the positive input (IN+), negative input (IN-), drive right leg output (DRL), and high-impedance state (off) of analog front-end 304, respectively. Each electrode is connected to the four outputs via four independent analog switches, and microcontroller 301 controls the on / off state of each analog switch via the I2C bus. At any given time, one electrode is connected to IN+, one to IN-, and one to DRL, while the remaining electrodes are in a high-impedance state. By configuring the switch matrix, three of the six electrodes can be selected as the positive, negative, and drive right leg electrodes, respectively.

[0048] The right leg drive circuit 305 includes an operational amplifier and a programmable feedback resistor network. The non-inverting input of the operational amplifier is connected to a reference voltage, the inverting input is connected to the common-mode nodes IN+ and IN- via resistors, and the output is connected to the DRL electrode. The programmable feedback resistor network consists of multiple resistors and analog switches. The microcontroller 301 controls the combination of switches to change the feedback resistor values, thereby adjusting the feedback gain of the right leg drive circuit. The feedback gain adjustment range is 20dB to 46dB.

[0049] The specific working process of the method of this invention is as follows:

[0050] like Figure 4 As shown, the system alternately performs ECG signal acquisition and impedance measurement in a time-division multiplexing manner. One working cycle T is 10ms, with the ECG acquisition time slot TECG being 9.2ms, the impedance measurement time slot TZ being 0.4ms, and the time slot switching interval being 0.2ms. During TECG, the analog switch matrix connects the selected electrode pair to the analog front-end 304, and the ECG signal is acquired via an instrumentation amplifier, a programmable gain amplifier, a programmable filter, and an analog-to-digital converter. At the start of TZ, the microcontroller 301 controls the analog switch to ground the input terminal of the ECG acquisition channel, and simultaneously connects the excitation output of the impedance measurement unit 303 to the electrode pair under test through a coupling capacitor; first, a 10kHz low-frequency excitation is injected, the response voltage is measured, and the low-frequency impedance ZLF is calculated; then, a 100kHz high-frequency excitation is injected, the response voltage is measured, and the high-frequency impedance ZHF is calculated; after the measurement is completed, the excitation is disconnected, and the ECG acquisition channel connection is restored.

[0051] Within each TZ time slot, the microcontroller 301 sequentially performs impedance measurements on the currently operating main electrode pair and all redundant electrode pairs. For each electrode pair, it calculates the resistive component Rs and capacitive component Cs of the low-frequency impedance, and the resistive component Rh and capacitive component C_h of the high-frequency impedance. At low frequencies, the capacitive reactance is relatively large, and the impedance mainly reflects the resistive characteristics of the stratum corneum and sweat; at high frequencies, the capacitive reactance decreases, and the impedance simultaneously reflects the capacitive characteristics of the electrode-skin interface and changes in contact area.

[0052] The microcontroller 301 classifies the contact state based on the impedance measurement results. The preset drying impedance threshold Zhigh is set to 500kΩ, the saturation impedance threshold Zlow is set to 20kΩ, the rate of change threshold R_th is set to 50kΩ / s, and the jump threshold is set to 100kΩ.

[0053] (1) Dry steady state: The contact impedance amplitude of the current working electrode pair is greater than 500kΩ, and the absolute value of the rate of change for three consecutive measurement cycles is less than 50kΩ / s. At this time, the electrode has good contact with the skin but the interface is dry, and the quality of the ECG signal is good. The microcontroller 301 sets the gain of the programmable gain amplifier 3041 to 24 times, the bandwidth of the programmable filter 3042 to 0.5Hz~100Hz, the sampling rate of the analog-to-digital converter 3043 to 500Hz and the resolution to 24 bits, and the feedback gain of the right leg drive to 26dB.

[0054] (2) Sweating transition period: The contact impedance amplitude is between 20kΩ and 500kΩ, and the absolute value of the rate of change is greater than 50kΩ / s. At the same time, the low-frequency resistance component Rs decreases rapidly while the high-frequency capacitance component Ch remains basically unchanged (the change is less than 10%). This indicates that sweat is wetting the stratum corneum but has not yet formed a continuous liquid film, which is the initial stage of sweating. The microcontroller 301 reduces the high-pass cutoff frequency from 0.5Hz to 0.05Hz to suppress baseline drift caused by sweat, and increases the gain from 24 times to 48 times to compensate for the signal amplitude attenuation caused by sweat shunting, while keeping the bandwidth and sampling rate unchanged.

[0055] (3) Sweat saturation state: The contact impedance amplitude is less than 20kΩ and lasts for more than 5 seconds, and both the low-frequency resistance component Rs and the high-frequency resistance component Rh decrease significantly. This indicates that sweat has formed a continuous liquid film on the electrode surface, and a bypass short circuit has occurred between adjacent electrodes. The microcontroller 301 first queries the impedance values ​​of all redundant electrode pairs in the impedance matrix, selects the pair of redundant electrodes with the best contact impedance, and switches the ECG acquisition channel to this redundant electrode pair through the analog switch matrix 4; at the same time, it increases the feedback gain of the right leg drive from 26dB to 40dB to enhance the common-mode rejection capability; after the switching is completed, the contact state of the new electrode pair is re-evaluated.

[0056] (4) Motion-induced contact loss state: The contact impedance jumps more than 100kΩ between adjacent measurement cycles, and both the low-frequency resistance component and the high-frequency capacitance component fluctuate rapidly. This indicates that there has been a momentary detachment or relative displacement between the electrode and the skin. The microcontroller 301 narrows the filter bandwidth to 0.5Hz~40Hz to suppress high-frequency motion artifacts, reduces the sampling rate to 250Hz to reduce redundant data, and simultaneously queries the impedance matrix to select the electrode pair with the lowest and most stable contact impedance for switching; the analog-to-digital converter resolution is increased to 24 bits to maintain the effective number of bits under narrow-band conditions.

[0057] Based on the above classification and compensation, this invention also performs predictive compensation. The microcontroller 301 records the impedance value sequence of the most recent 20 impedance measurement cycles using a sliding window method, performs linear fitting on the impedance drop segment during the sweating transition period, and obtains the impedance change rate k (in kΩ / s). Based on the difference between the current impedance Zcurrent and the saturation threshold Zlow, divided by the change rate k, the remaining time to reach sweat saturation, tpred = (Zcurrent - Zlow) / |k|, is estimated. When tpred is less than 3 seconds, the microcontroller 301 selects the optimal redundant electrode pair from the impedance matrix in advance and completes the switching operation, ensuring that the switching operation is completed before sweat saturation causes severe signal degradation. After switching, the original main electrode is marked as "to be recovered," and the system continuously monitors its impedance in subsequent impedance measurement time slots. When the impedance of the original main electrode recovers to the dry stable state range and remains there for more than 10 seconds, it is reintroduced into the available electrode pool.

[0058] In addition, the microcontroller 301 dynamically adjusts the impedance measurement frequency based on the motion intensity detected by the accelerometer 307. When the acceleration variance is less than 0.1g, it is determined to be a resting state, and the impedance measurement cycle is extended to 100ms (measured once every 10 ECG cycles) to reduce power consumption; when the acceleration variance is greater than 0.5g, it is determined to be high-intensity exercise, and the impedance measurement cycle is shortened to 10ms (measured in every cycle) to speed up the state response.

[0059] When using the smart clothing of this invention, the wearer fastens the main control module 3 to the module interface 205 at the back of the clothing neckline, and the textile electrode array 2 is attached to the chest skin after wearing the clothing. After the system starts, the first main electrode 201a and the second main electrode 201b are selected by default as the positive and negative poles for ECG acquisition, and the third main electrode 201c is used as the driving electrode for the right leg. After exercise begins, the system continuously monitors impedance, classifies the state, and performs compensation according to the above method. In a 45-minute running test, the wearer began to sweat at about 15 minutes. The system detected that the impedance dropped rapidly from about 600kΩ, classified it as the sweating transition period, and automatically reduced the high-pass cutoff frequency and increased the gain. At about 20 minutes, the impedance dropped to 15kΩ and entered the sweat saturation state. The system completed the redundant electrode scan within the 0.4ms impedance measurement time slot and switched to the second redundant electrode 202b and the third redundant electrode 202c. The ECG signal returned to normal amplitude after the switch, and there was no perceptible interruption of ECG data during the entire switching process.

[0060] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for compensating electrocardiogram signals based on impedance feedback, characterized in that, Includes the following steps: S1. ECG signal acquisition and impedance measurement are performed alternately in a time-division multiplexing manner using a textile electrode array; during the impedance measurement time slot, an AC excitation signal with known amplitude and frequency is injected into at least one pair of textile electrodes, the response voltage at both ends of the electrodes is measured, and the contact impedance value of the skin-electrode interface is calculated. S2. Based on the amplitude, rate of change, and frequency response characteristics of the contact impedance, the skin-electrode contact state is classified into one of the following: dry stable state, sweating transition period, sweat saturation state, and motion-induced contact loss state. S3. Execute corresponding compensation strategies based on the classification results: When in a dry and stable state, use standard gain and broadband filtering; when in a sweating transition period, reduce the high-pass filter cutoff frequency to suppress baseline drift, and increase the instrumentation amplifier gain to compensate for signal attenuation caused by sweat shunting; when in a sweat-saturated state, switch to redundant electrode pairs and adjust the feedback gain of the right leg drive circuit; when in a motion-induced contact loss state, narrow the filter bandwidth and reduce the sampling rate, and switch to the redundant electrode pairs with the lowest contact impedance.

2. The ECG signal compensation method based on impedance feedback according to claim 1, characterized in that, In step S1, the AC excitation signal includes a low-frequency excitation signal and a high-frequency excitation signal. The frequency of the low-frequency excitation signal is 8kHz to 12kHz, and the frequency of the high-frequency excitation signal is 80kHz to 120kHz. The resistance component of the contact impedance is calculated based on the low-frequency excitation response, and the capacitance component of the contact impedance is calculated based on the high-frequency excitation response. When the resistance component decreases rapidly while the capacitance component remains basically unchanged, it is determined to be the sweating transition period. When both the resistance component and the capacitance component fluctuate rapidly, it is determined to be the motion loss state.

3. The ECG signal compensation method based on impedance feedback according to claim 1, characterized in that, In step S2, the classification method specifically includes: preset impedance thresholds Zhigh, Zlow, and a rate of change threshold Rth; when the contact impedance amplitude is greater than Zhigh and the absolute value of the rate of change is less than Rth, it is classified as a dry stable state; when the contact impedance amplitude is between Zlow and Zhigh and the absolute value of the rate of change is greater than Rth, it is classified as a sweating transition period; when the contact impedance amplitude is less than Zlow and the duration exceeds the set duration, it is classified as a sweat saturation state; when the jump in contact impedance amplitude between adjacent measurement cycles exceeds the set jump threshold, it is classified as a motion-induced contact loss state.

4. The ECG signal compensation method based on impedance feedback according to claim 1, characterized in that, In step S3, the method for switching redundant electrode pairs is as follows: the textile electrode array includes at least three main electrodes and redundant electrodes arranged adjacent to each main electrode. Each electrode is connected to the signal acquisition channel through an analog switch matrix. In each impedance measurement time slot, the contact impedance of all available electrode pairs is measured sequentially to establish an impedance matrix. When the contact impedance of any electrode in the main electrode pair deteriorates to the trigger threshold, the redundant electrode pair with the best contact impedance is selected from the impedance matrix to replace the deteriorated main electrode, and the selection state of the analog switch matrix is ​​updated.

5. The ECG signal compensation method based on impedance feedback according to claim 1, characterized in that, In step S3, the feedback gain adjustment method of the right leg driving circuit is as follows: calculate the common-mode voltage between the current working electrode pair in real time, and adjust the feedback resistance value of the right leg driving amplifier in a step manner according to the deviation between the common-mode voltage and the target common-mode voltage; when sweat saturation causes the common-mode rejection ratio to decrease, increase the feedback gain to enhance the common-mode rejection capability; when the dry and stable state is reached, decrease the feedback gain to avoid saturation of the right leg driving circuit.

6. The ECG signal compensation method based on impedance feedback according to claim 1, characterized in that, It also includes a prediction compensation step: continuously record the trajectory of contact impedance change over time, and when the impedance is detected to enter the sweating transition period and the impedance change rate shows a continuous downward trend, predict the time to reach the sweat saturation state based on the impedance change rate extrapolation, and actively switch to the redundant electrode pair with a preset advance amount before the prediction is reached, so that the switching operation is completed before the signal is severely degraded.

7. The ECG signal compensation method based on impedance feedback according to claim 1, characterized in that, In step S1, the time ratio of the ECG signal acquisition time slot to the impedance measurement time slot is (8~12):1, the duration of each impedance measurement time slot does not exceed 0.5ms, and the impedance measurement time slot is inserted between two adjacent ECG sampling intervals; within the impedance measurement time slot, the input terminal of the ECG acquisition channel is grounded through an analog switch, and the excitation signal is injected into the electrode through an independent coupling capacitor.

8. A smart garment, characterized in that, This includes the garment body, textile electrode array, main control module, and analog switch matrix; The textile electrode array includes at least three main textile electrodes and at least three redundant textile electrodes. The main textile electrodes are respectively located on the garment body at the intersection of the right subclavian, left subclavian, and left anterior axillary lines with the lower edge of the costal arch (approximately the 6th intercostal space). Each redundant textile electrode is located 2cm to 4cm to the side of the corresponding main textile electrode. Each textile electrode is connected to the module interface on the garment body through conductive yarns woven into the fabric. The main control module is detachably installed on the garment body and electrically connected to the textile electrode array through the module interface. The main control module includes: a signal acquisition unit for acquiring electrocardiogram signals; an impedance measurement unit for generating AC excitation signals and measuring response voltage to calculate contact impedance; an analog front end including an instrumentation amplifier, a programmable gain amplifier, a programmable filter, and an analog-to-digital converter; a right leg drive circuit; and a microcontroller for executing the method of any one of claims 1 to 7. The analog switch matrix is ​​positioned between the textile electrode array and the analog front end. Each textile electrode is connected to the positive input terminal, negative input terminal, or right leg drive output terminal of the analog front end via an analog switch. The microcontroller selects the currently working electrode pair and the right leg drive electrode by controlling the on / off state of the analog switch matrix.

9. The smart clothing according to claim 8, characterized in that, The main control module also includes a Bluetooth communication unit and an accelerometer. The accelerometer is used to collect human motion acceleration. The microcontroller adjusts the impedance measurement frequency according to the motion intensity output by the accelerometer: when the motion intensity is detected to increase, the impedance measurement frequency is increased to speed up the state response speed; when the motion intensity is detected to decrease, the impedance measurement frequency is decreased to reduce power consumption.

10. The smart clothing according to claim 8, characterized in that, Each textile electrode in the textile electrode array has a ring reference electrode at its edge. The ring reference electrode is concentrically arranged with the corresponding textile electrode and is insulated from it. During impedance measurement, an excitation signal is injected into the textile electrode, and the response voltage is detected from the ring reference electrode, so that the current path is concentrated in the skin area directly below the textile electrode.

Images