Sub-band signal decomposition method and portable wearable device based on flexible electronic skin

Abstract

The invention relates to a subband signal decomposition method and a portable wearable device based on flexible electronic skin, and relates to the technical field of signal processing. In the sub-band signal decomposition method, whether a current motion mode is a strenuous motion mode is judged based on an acceleration signal, acceleration signal energy and an impedance change rate, if yes, a physiological signal needing to be decomposed is decomposed into a first preset number of sub-bands, and if not, the physiological signal needing to be decomposed is decomposed into a second preset number of sub-bands. Wherein the acceleration signal energy is calculated based on the acceleration of each sampling point in the x, y and z directions in the first preset integral window. The method can comprehensively reflect the quantity of the cumulative effect of the acceleration in a period of time, and can more comprehensively reflect the intensity and duration of motion, so that sub-band division can be more accurately adjusted according to the change of energy, the frequency and intensity change of motion artifacts can be tracked in real time, the characteristics of the motion artifacts can be better captured, the coupling noise can be inhibited, and the accuracy of motion artifacts can be improved. And measurement errors are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of signal processing, and particularly to a sub-band signal decomposition method and a portable wearable device based on a flexible electronic skin. Background Art

[0002] The frequency ranges of motion artifacts, electromagnetic interference, and contact impedance fluctuations of a flexible electronic skin are key indicators affecting its performance. Among them, motion artifacts are mainly caused by the micro-movement of the skin-electrode interface of a living organism, and the acceleration amplitude is directly related to the exercise intensity. Statistical analysis of the exercise acceleration of healthy adults: slow walking (1 m / s): the mean range of triaxial acceleration is 0.2 - 0.5 g (where g is the gravitational acceleration constant); fast running (3 m / s): the acceleration peak can reach 1.8 g; jumping (vertical height 20 cm): the acceleration peak is as high as 2.0 - 3.0 g. An acceleration threshold of 1.5 g is often used to suppress motion artifacts, that is, to cover the sensitive responses of high-artifact scenarios such as fast running (1.6 g) and above high-intensity exercises, and also to avoid false triggering of low-intensity exercises (such as 0.8 g of slow walking).

[0003] For a flexible electronic skin, the rapid change of the skin-electrode contact impedance of a living organism is also one of the main sources of motion artifacts. The conductive network of the flexible electrode breaks and reorganizes during dynamic deformation, resulting in discontinuous carrier migration paths. Micro-cracks are generated at the multi-layer structure interface due to the difference in Young's modulus, further exacerbating impedance jumps. Through research: static contact: impedance volatility < 1 kΩ / s; mild exercise (such as arm swing): impedance change rate 2 - 4 kΩ / s; strenuous exercise (such as jumping): impedance change rate can reach 6 - 10 kΩ / s. The impedance change rate of 5 kΩ / s corresponds to a significant drop critical point of SNR < 24 dB, and a compensation algorithm needs to be immediately triggered to suppress artifacts.

[0004] Therefore, in strenuous exercise scenarios, the coupled noise brought by the above several sources will cause great measurement errors and affect the practical application value. Summary of the Invention

[0005] The technical problem to be solved by this application is to provide a sub-band signal decomposition method and a portable wearable device based on a flexible electronic skin, which have the characteristics of better suppressing coupled noise and reducing measurement errors.

[0006] In a first aspect, in one embodiment, a sub-band signal decomposition method is provided, including: Collect physiological signals, acceleration signals, and contact impedance signals based on the same time window that need to be decomposed, and calculate the acceleration signal energy and impedance change rate; Based on the acceleration signal, the acceleration signal energy, and the impedance change rate, determine whether the current motion mode is a strenuous exercise mode. If so, decompose the physiological signal to be decomposed into a first preset number of subbands; if not, decompose the physiological signal to be decomposed into a second preset number of subbands. Among them, the first preset number is greater than the second preset number; The calculation of the acceleration signal energy includes: calculating the acceleration signal energy based on the accelerations of each sampling point in the x, y, and z directions within a first preset integration window; The determination of whether the current motion mode is a strenuous exercise mode based on the acceleration signal, the acceleration signal energy, and the impedance change rate includes: Determine whether the acceleration peak value is greater than a preset acceleration threshold, whether the acceleration signal energy is greater than a preset energy threshold, or whether the impedance change rate is greater than a preset change rate threshold. If so, determine that the current motion model is a strenuous exercise mode.

[0007] In one embodiment, the calculation of the acceleration signal energy based on the accelerations of each sampling point in the x, y, and z directions within a preset integration window includes: Among them, represents the acceleration signal energy, 、 and respectively represent the accelerations of the i-th sampling point in the x direction, y direction, and z direction. m represents the total number of sampling points within the first preset integration window, n represents the current time point, and n - m ≤ i ≤ n.

[0008] In one embodiment, the determination of whether the acceleration peak value is greater than a preset acceleration threshold, whether the acceleration signal energy is greater than a preset energy threshold, or whether the impedance change rate is greater than a preset change rate threshold includes: Determine whether it satisfies or or , where ɡ is the gravitational acceleration constant, is the preset acceleration threshold, is the preset acceleration signal energy threshold, is the preset change rate threshold; represents taking the maximum value among the accelerations of all sampling points in the x direction, y direction, and z direction.

[0009] In one embodiment, the first preset number is 16, and the second preset number is 8.

[0010] In one embodiment, the method further includes: for any sub-band, determining in real time whether the sub-band error energy is greater than a preset first error energy threshold. If so, all sub-bands are split into sub-bands with twice the number to ensure frequency band continuity.

[0011] In one embodiment, for any sub-band, determining in real time whether the sub-band error energy is greater than a preset first error energy threshold includes: For any sub-band, obtaining the actual output signal and the desired output signal of each sampling point within a second preset integration window; Based on the actual output signal and the desired output signal of each sampling point, calculating the error signal of each sampling point; Calculating the sub-band error energy based on the error signals of each sampling point; Determining whether the sub-band error energy is greater than a preset first error energy threshold.

[0012] In one embodiment, calculating the error signal of each sampling point based on the actual output signal and the desired output signal of each sampling point includes: Wherein, , and represent the error signal, the actual output signal, and the desired output signal of the jth sampling point in the kth sub-band; Calculating the sub-band error energy based on the error signals of each sampling point includes: Wherein, represents the sub-band error energy of the kth sub-band, N represents the total number of sampling points within the second preset integration window, and 0 ≤ j ≤ N - 1; Determining whether the sub-band error energy is greater than a preset first error energy threshold includes: determining whether the sub-band error energy satisfies ; Wherein, represents the noise power of the kth sub-band, represents the sub-band error energy from the 1st to the Mth sub-bands, and M represents the total number of sub-bands, where 1 ≤ k ≤ M.

[0013] In one embodiment, the method further includes: determining in real time whether the sub-band error energies of all sub-bands for a continuous preset number of frames threshold are all less than a preset second error energy threshold. If so, pairwise merging of adjacent sub-bands is performed on all sub-bands.

[0014] In one embodiment, for any sub-band of any frame, determining whether the sub-band error energy is less than a preset second error energy threshold includes: determining whether the sub-band error energy satisfies ; wherein, represents the sub-band error energy of the k-th sub-band, represents the noise power of the k-th sub-band.

[0015] In a second aspect, an embodiment provides a portable wearable device based on a flexible electronic skin, which is characterized by comprising a physiological signal acquisition module, an accelerometer, a contact impedance sensor, at least one processor and at least one readable storage medium; the physiological signal acquisition module is used for acquiring physiological signals of a user, the accelerometer is used for acquiring acceleration signals, and the contact impedance sensor is used for acquiring contact impedance signals; a program is stored in the at least one readable storage medium, and the program can be loaded and executed by the at least one processor to perform any one of the sub-band signal decomposition methods in the above embodiments; the at least one processor is used for acquiring physiological signals, acceleration signals and contact impedance signals, and loading and executing any one of the sub-band signal decomposition methods in the above embodiments to obtain decomposed sub-band signals, and further processing the sub-band signals to capture the characteristics of motion artifacts and suppress coupled noise; wherein, the physiological signal is used as the physiological signal to be decomposed.

[0016] The beneficial effects of the present invention are as follows: In the sub-band signal decomposition method, based on the acceleration signal, the acceleration signal energy and the impedance change rate, it is determined whether the current motion mode is a strenuous motion mode. If so, the physiological signal to be decomposed is decomposed into a first preset number of sub-band numbers. If not, the physiological signal to be decomposed is decomposed into a second preset number of sub-band numbers. In this way, the quantity that can comprehensively reflect the cumulative effect of acceleration over a period of time can be obtained. Compared with the acceleration peak value, it can more comprehensively reflect the intensity and duration of the motion, so as to more precisely adjust the sub-band division according to the change of energy, and track the frequency and intensity changes of motion artifacts in real time, so that the characteristics of motion artifacts can be better captured, the coupled noise can be suppressed, and the measurement error can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a schematic structural diagram of a portable wearable device based on a flexible electronic skin according to an embodiment of the present application; Figure 2 is a schematic flowchart of a method for sub-band signal decomposition according to an embodiment of the present application; Figure 3 is a schematic flowchart of a method for real-time determining whether the sub-band error energy is greater than a preset first error energy threshold according to an embodiment of the present application.

[0018] In the illustration: 01 is a physiological signal acquisition module, 02 is an accelerometer, 03 is a contact impedance sensor, 04 is a processor, and 05 is a readable storage medium. Specific embodiments

[0019] The present invention will be further described in detail below in conjunction with the accompanying drawings through specific embodiments. Similar elements in different embodiments are labeled with related similar element numbers. In the following embodiments, many detailed descriptions are provided to enable a better understanding of the present application. However, those skilled in the art can easily recognize that some of the features can be omitted in different situations, or can be replaced by other elements, materials, or methods. In some cases, some operations related to the present application are not shown or described in the specification to avoid overwhelming the core part of the present application with excessive descriptions. For those skilled in the art, it is not necessary to describe these related operations in detail, and they can fully understand the related operations based on the descriptions in the specification and general technical knowledge in the art.

[0020] In addition, the features, operations, or characteristics described in the specification can be combined in any appropriate manner to form various embodiments. At the same time, the steps or actions in the method description can also be reordered or adjusted in an obvious manner by those skilled in the art. Therefore, the various sequences in the specification and drawings are only for clearly describing a certain embodiment and do not mean that they are necessary sequences unless it is stated that a certain sequence must be followed.

[0021] The numbers assigned to the components herein, such as "first", "second", etc., are only used to distinguish the objects described and do not have any sequential or technical meaning.

[0022] For the convenience of explaining the inventive concept of the present application, the sub-band decomposition technology is briefly described below.

[0023] For the frequency-domain signal to be decomposed, in the current technology, it is mostly decomposed into a fixed number of sub-bands. However, this decomposition method is difficult to adapt to the change in the movement amplitude to dynamically adjust the number of sub-bands. Therefore, it is difficult to better capture the characteristics of motion artifacts and thus better suppress coupling noise. Thus, in the current technology, the number of sub-bands is dynamically adjusted based on the acceleration peak value to better capture the characteristics of motion artifacts and suppress coupling noise.

[0024] However, the applicant found in the research that the peak acceleration cannot well reflect the duration of motion and the overall energy distribution, and will miss some motion artifacts with low energy but continuous existence, resulting in misjudgment or incomplete processing when suppressing coupled noise, especially for those non-periodic or strongly varying coupled noises. Moreover, since the peak acceleration is more susceptible to instantaneous interference, the sub-band division may be frequently adjusted due to occasional high peak accelerations. Such unstable sub-band division may affect the processing effect of motion artifacts and coupled noise. These have all become technical problems to be solved.

[0025] In view of this, the present application provides a sub-band signal decomposition method and a portable wearable device based on a flexible electronic skin. In the sub-band signal decomposition method, based on the acceleration signal, the acceleration signal energy, and the impedance change rate, it is determined whether the current motion mode is a strenuous motion mode. If so, the physiological signal to be decomposed is decomposed into a first preset number of sub-bands. If not, the physiological signal to be decomposed is decomposed into a second preset number of sub-bands. In this way, the quantity that can comprehensively reflect the cumulative effect of acceleration over a period of time can be obtained. Compared with the peak acceleration, it can more comprehensively reflect the intensity and duration of motion, so as to more accurately adjust the sub-band division according to the energy change, track the frequency and intensity changes of motion artifacts in real time, so that the characteristics of motion artifacts can be better captured, the coupled noise can be suppressed, and the measurement error can be reduced.

[0026] To better understand the solution of the present application, the portable wearable device based on a flexible electronic skin will be introduced first below.

[0027] A portable wearable device based on a flexible electronic skin provided by an embodiment of the present application, please refer to Figure 1 , including a physiological signal acquisition module 01, an accelerometer 02, a contact impedance sensor 03, at least one processor 04, and at least one readable storage medium 05.

[0028] Among them, the physiological signal acquisition module 01 is used to acquire the physiological signal of the user. In one embodiment, the acquired physiological signal includes at least one of physiological signals such as photoplethysmogram signal (PPG), electrocardiogram signal (ECG), and electromyogram signal (EMG).

[0029] The accelerometer 02 is used to acquire the acceleration signal. In one embodiment, the accelerometer 02 uses an integrated MEMS triaxial accelerometer, and the sampling frequency is 200 Hz.

[0030] The contact impedance sensor 03 is used to acquire the contact impedance signal. In one embodiment, the contact impedance sensor 03 uses a set piezoresistive contact impedance sensing layer.

[0031] Those skilled in the art can understand that a synchronous clock can be added to ensure that the timestamps of the collected physiological signals, acceleration signals, and contact impedance signals are consistent, avoiding errors in noise source localization caused by asynchrony.

[0032] At least one readable storage medium 05 stores a program that can be loaded and executed by at least one processor 04 to include the sub-band signal decomposition method in the embodiments of the present application. The at least one processor 04 is used to acquire physiological signals, acceleration signals, and contact impedance signals, and load and execute the sub-band signal decomposition method in the embodiments of the present application to obtain the decomposed sub-band signals, and further process the sub-band signals to capture the characteristics of motion artifacts and suppress coupled noise; wherein, the physiological signal is used as the physiological signal to be decomposed.

[0033] A portable wearable device based on a flexible electronic skin that can be applied to any of the above embodiments. For the sub-band signal decomposition method provided by the embodiments of the present application, please refer to Figure 2 , including: Step S10, collect physiological signals, acceleration signals, and contact impedance signals based on the same time window that need to be decomposed, and calculate the acceleration signal energy and impedance change rate.

[0034] In one embodiment, calculating the acceleration signal energy includes: calculating the acceleration signal energy based on the accelerations of each sampling point in the x, y, and z directions within a first preset integration window. In one embodiment, it can be expressed as: Among them, represents the acceleration signal energy, , and respectively represent the accelerations of the i-th sampling point in the x direction, y direction, and z direction, m represents the total number of sampling points within the first preset integration window, n represents the current time point, and n - m ≤ i ≤ n.

[0035] In one embodiment, m = 40.

[0036] In one embodiment, the impedance change rate can be expressed as . Among them, represents the impedance change value, represents the time interval.

[0037] Step S20, based on the acceleration signal, acceleration signal energy, and impedance change rate, determine whether the current motion mode is a strenuous motion mode. If so, go to step S30; if not, go to step S40.

[0038] Based on the acceleration signal, the acceleration signal energy, and the impedance change rate, determine whether the current motion mode is a strenuous exercise mode, including: determining whether the acceleration peak value is greater than a preset acceleration threshold, whether the acceleration signal energy is greater than a preset energy threshold, or whether the impedance change rate is greater than a preset change rate threshold, which can be expressed as: determining whether the following is satisfied or or , where \(g\) is the gravitational acceleration constant, is the preset acceleration threshold, is the preset acceleration signal energy threshold, is the preset change rate threshold; represents taking the maximum value among the accelerations in the x-direction, y-direction, and z-direction at all sampling points. If so, that is, if any one of the conditions is satisfied, then determine that the current motion model is a strenuous exercise mode. represents taking the maximum value among the accelerations in the x-direction, y-direction, and z-direction at all sampling points.

[0039] Step S30, decompose the physiological signal to be decomposed into a first preset number of subbands.

[0040] Step S40, divide the physiological signal to be decomposed into a second preset number of subbands. Wherein the first preset number is greater than the second preset number.

[0041] In one embodiment, the first preset number is 16, that is, in the strenuous exercise mode, the physiological signal to be decomposed is decomposed into 16 subbands. The second preset number is 8, that is, in the non-strenuous exercise mode, the physiological signal to be decomposed is decomposed into 8 subbands. Thus, in the initial state of the non-strenuous exercise mode, the initial value of the number of subbands is 8. If it is determined to be the strenuous exercise mode, the number of decomposed subbands is 16.

[0042] In one implementation, the subbands cover 0.1 Hz - 100 Hz, are decomposed into 16 subbands during strenuous exercise, each subband having a width of 6.25 Hz, and are decomposed into 8 subbands during non-strenuous exercise, each subband having a width of 25 Hz.

[0043] In the above embodiment, through the identification of impedance fluctuations, the distortion of the ECG signal can be effectively identified. Through the detection of the acceleration peak value, the situation where the integration window masks instantaneous strenuous exercise can be avoided. Through the detection of the acceleration signal energy, the quantity reflecting the cumulative effect of acceleration over a period of time can be comprehensively reflected. Compared with the acceleration peak value, it can more comprehensively reflect the intensity and duration of the exercise. Using the acceleration signal energy for dynamic decomposition can more precisely adjust the subband division according to the change of energy, and track the frequency and intensity changes of motion artifacts in real time. Capture motion artifacts more meticulously, and then remove them more effectively.

[0044] However, the applicant found in the research that since the peak acceleration is more susceptible to instantaneous interference, the sub-band division may be frequently adjusted due to occasional high acceleration peaks. Therefore, the presence of peak acceleration detection may still lead to misjudgment or incomplete processing when suppressing coupling noise. To solve this technical problem, a further solution is proposed in the embodiments of the present application.

[0045] In a further solution, in one embodiment, for any sub-band, it is determined in real time whether the sub-band error energy is greater than a preset first error energy threshold. If so, all sub-bands are split into twice the number of sub-bands to ensure frequency band continuity.

[0046] In one embodiment, for any sub-band, it is determined in real time whether the sub-band error energy is greater than a preset first error energy threshold. Please refer to Figure 3 , including: Step S100, for any sub-band, obtain the actual output signal and the desired output signal of each sampling point within the second preset integration window.

[0047] In one embodiment, the second preset integration window includes 10 sampling points.

[0048] Step S200, based on the actual output signal and the desired output signal of each sampling point, calculate the error signal of each sampling point.

[0049] In one embodiment, step S200 can be expressed as: Wherein, , and represent the error signal, the actual output signal, and the desired output signal of the jth sampling point in the kth sub-band.

[0050] Step S300, calculate the sub-band error energy based on the error signals of each sampling point.

[0051] In one embodiment, step S300 can be expressed as: Wherein, represents the sub-band error energy of the kth sub-band, N represents the total number of sampling points within the second preset integration window, and 0 ≤ j ≤ N - 1. If the second preset integration window includes 10 sampling points, then N = 10.

[0052] In some embodiments, the desired output signal can be an artifact-free signal collected by a reference electrode or a predicted signal based on historical data.

[0053] Step S400: Determine whether the sub-band error energy is greater than a preset first error energy threshold.

[0054] In one embodiment, step S400 can be expressed as determining whether the sub-band error energy satisfies ; where represents the noise power of the k-th sub-band, represents the sub-band error energy from the 1st to the M-th sub-band, M represents the total number of sub-bands, and 1 ≤ k ≤ M.

[0055] In one embodiment, the noise power of the k-th sub-band can be obtained in the following manner: First, based on the acceleration signal energy and the change rate of contact impedance, construct a polynomial regression model to calculate the global noise power , which can be expressed as: ; where is the reference value of the contact impedance in the non-vigorous exercise state, represents the impedance at the current moment, , and are fitting parameters. In one embodiment, = 0.46, = 0.07, = 0.12.

[0056] For any sub-band k, perform a fast Fourier transform (FFT) to obtain the frequency-domain energy spectrum of sub-band k, detect the frequency points that satisfy , and record them as the noise frequency frequency-domain energy set. Among them, is the noise floor of sub-band k, which is obtained by calibrating the non-vigorous exercise state data.

[0057] Secondly, calculate the noise energy ratio of sub-band k based on the noise frequency set of each sub-band k, which can be expressed as: where represents the frequency-domain energy of the r-th noise frequency in the k-th sub-band, represents the number of noise frequencies in the k-th sub-band, and 1 ≤ r ≤ ; represents the frequency-domain energy of the q-th noise frequency in the p-th sub-band, represents the number of noise frequencies in the p-th sub-band, and 1 ≤ q ≤ , and 1 ≤ p ≤ M.

[0058] The noise power of sub-band k can then be obtained. .

[0059] In one embodiment, the maximum value of the number of decomposed sub-bands is 16. That is, regardless of whether there is a sub-band error energy greater than a preset first error energy threshold, if the number of current sub-bands has reached 16, no further sub-band splitting is performed.

[0060] Based on the above sub-band splitting method, the situation of misjudgment or incomplete processing during the suppression of coupled noise can be reduced, and the measurement error can be reduced.

[0061] In a further solution, in one embodiment, it is determined in real time whether the sub-band error energies of all sub-bands for a continuous preset number of frames threshold all satisfy being less than a preset second error energy threshold. If so, all sub-bands are merged pairwise with adjacent sub-bands.

[0062] In one embodiment, for any sub-band of any frame, determining whether the sub-band error energy is less than a preset second error energy threshold includes: determining whether the sub-band error energy satisfies ; wherein, represents the sub-band error energy of the k-th sub-band, represents the noise power of the k-th sub-band.

[0063] In one embodiment, it is determined in real time whether the sub-band error energies of all sub-bands for 5 consecutive frames all satisfy being less than a preset second error energy threshold.

[0064] Based on the above sub-band splitting method, the situation of misjudgment or incomplete processing during the suppression of coupled noise can be further reduced, and the measurement error can be reduced.

[0065] In one embodiment of the present application, a computer-readable storage medium is provided, and a program is stored on the storage medium. The stored program includes methods that can be loaded and processed by a processor in any of the above embodiments.

[0066] Those skilled in the art can understand that all or part of the functions of the various methods in the above embodiments can be implemented in a hardware manner or in a computer program manner. When all or part of the functions in the above embodiments are implemented in a computer program manner, the program can be stored in a computer-readable storage medium, and the storage medium can include: read-only memory, random access memory, magnetic disk, optical disk, hard disk, etc. The above functions can be achieved by a computer executing the program. For example, the program is stored in the memory of the device, and when the processor executes the program in the memory, the above all or part of the functions can be achieved. In addition, when all or part of the functions in the above embodiments are implemented in a computer program manner, the program can also be stored in a storage medium such as a server, another computer, magnetic disk, optical disk, flash drive or mobile hard disk, and saved to the memory of the local device by downloading or copying, or the system of the local device is updated. When the processor executes the program in the memory, all or part of the functions in the above embodiments can be achieved.

[0067] The above uses specific examples to elaborate on the present invention, which is only used to help understand the present invention and is not intended to limit the present invention. For those skilled in the art of the present invention, according to the idea of the present invention, several simple deductions, deformations or substitutions can also be made.

Claims

1. A subband signal decomposition method, characterized in that Including: Collect physiological signals, acceleration signals, and contact impedance signals based on the same time window that need to be decomposed, and calculate the acceleration signal energy and impedance change rate; Based on the acceleration signal, acceleration signal energy, and impedance change rate, determine whether the current motion mode is a strenuous exercise mode. If so, decompose the physiological signal to be decomposed into a first preset number of sub-bands. If not, decompose the physiological signal to be decomposed into a second preset number of sub-bands; wherein, the first preset number is greater than the second preset number; The calculation of the acceleration signal energy includes: calculating the acceleration signal energy based on the accelerations of each sampling point in the x, y, and z directions within a first preset integration window; The determination of whether the current motion mode is a strenuous exercise mode based on the acceleration signal, acceleration signal energy, and impedance change rate includes: Determine whether the acceleration peak value is greater than a preset acceleration threshold, whether the acceleration signal energy is greater than a preset energy threshold, or whether the impedance change rate is greater than a preset change rate threshold. If so, determine that the current motion model is a strenuous exercise mode.

2. The sub-band signal decomposition method according to claim 1, characterized in that, The calculation of the acceleration signal energy based on the accelerations of each sampling point in the x, y, and z directions within a preset integration window includes: Among them, represents the acceleration signal energy, , and respectively represent the accelerations of the i-th sampling point in the x-direction, y-direction, and z-direction. m represents the total number of sampling points within the first preset integration window, n represents the current time point, and n - m ≤ i ≤ n.

3. The sub-band signal decomposition method according to claim 2, characterized in that The determination of whether the acceleration peak value is greater than a preset acceleration threshold, whether the acceleration signal energy is greater than a preset energy threshold, or whether the impedance change rate is greater than a preset change rate threshold includes: Determine whether it meets or or , where ɡ is the gravitational acceleration constant, is the preset acceleration threshold, is the preset acceleration signal energy threshold, is the preset rate of change threshold; represents taking the maximum value among the accelerations in the x-direction, y-direction, and z-direction at all sampling points.

4. The sub-band signal decomposition method according to claim 1, characterized in that, The first preset number is 16, and the second preset number is 8.

5. The sub-band signal decomposition method according to claim 1, wherein The method further includes: for any one sub-band, real-time determine whether the sub-band error energy is greater than a preset first error energy threshold. If so, split all sub-bands into twice the number of sub-bands to ensure frequency band continuity.

6. The sub-band signal decomposition method according to claim 5, characterized in that, The real-time determination of whether the sub-band error energy is greater than a preset first error energy threshold for any one sub-band includes: For any one sub-band, obtain the actual output signal and expected output signal of each sampling point within its second preset integration window; Based on the actual output signal and expected output signal of each sampling point, calculate the error signal of each sampling point; Based on the error signals of each sampling point, calculate the sub-band error energy; Determine whether the sub-band error energy is greater than a preset first error energy threshold.

7. The sub-band signal decomposition method according to claim 5, wherein The calculation of the error signal of each sampling point based on the actual output signal and expected output signal of each sampling point includes: Among them, , and represent the error signal, actual output signal, and desired output signal of the j-th sampling point of the k-th sub-band; The calculation of the sub-band error energy based on the error signals of each sampling point includes: Among them, represents the sub-band error energy of the k-th sub-band, N represents the total number of sampling points within the second preset integration window, and 0 ≤ j ≤ N - 1; The determination of whether the sub-band error energy is greater than a preset first error energy threshold includes: determining whether the sub-band error energy satisfies ； Among them, represents the noise power of the k-th subband, represents the subband error energy of the first to M subbands, where M represents the total number of subbands and 1 ≤ k ≤ M.

8. The sub-band signal decomposition method according to claim 1, characterized in that The method further includes: real-time determine whether the sub-band error energies of all sub-bands for a continuous preset number of frames threshold are all less than a preset second error energy threshold. If so, perform pairwise merging of adjacent sub-bands for all sub-bands.

9. The sub-band signal decomposition method according to claim 8, wherein, For any one sub-band of any one frame, determining whether the sub-band error energy is less than a preset second error energy threshold includes: determining whether the sub-band error energy satisfies ； Among them, represents the subband error energy of the k-th subband, represents the noise power of the k-th subband.

10. A portable wearable device based on a flexible electronic skin, characterized in that, It includes a physiological signal acquisition module, an accelerometer, a contact impedance sensor, at least one processor and at least one readable storage medium; the physiological signal acquisition module is used to acquire the physiological signals of the user, the accelerometer is used to acquire acceleration signals, and the contact impedance sensor is used to acquire contact impedance signals; the at least one readable storage medium stores a program, and the program can be loaded and executed by the at least one processor to perform the sub-band signal decomposition method described in any one of claims 1 to 9; the at least one processor is used to obtain physiological signals, acceleration signals and contact impedance signals, and load and execute the sub-band signal decomposition method described in any one of claims 1 to 9 to obtain the decomposed sub-band signals, and further process the sub-band signals to capture the characteristics of motion artifacts and suppress coupled noise; wherein, the physiological signal is used as the physiological signal to be decomposed.

Images