Low-noise biopotential signal acquisition system and processing method

Abstract

The invention belongs to the technical field of signal acquisition and processing, and discloses a low-noise biopotential signal acquisition system and a low-noise biopotential signal processing method. An original bioelectricity signal of a human body is collected through an electrode array, and analog-to-digital conversion is carried out after the signal is processed by a multi-stage self-adaptive filtering and amplifying circuit. A digital signal is sequentially subjected to adaptive wavelet transform denoising and empirical mode decomposition to obtain a multi-level intrinsic mode function set, an intrinsic mode function of a characteristic frequency band is extracted from the multi-level intrinsic mode function set, and the signal is reconstructed. Introducing a dual noise reduction mechanism combining adaptive wavelet transform and empirical mode decomposition; a signal quality real-time evaluation system is established, and system parameters including the gain of a variable gain amplifier and the bandwidth of a dynamic band-pass filter are dynamically adjusted through closed-loop feedback. The self-adaptive optimization of the whole biopotential signal acquisition process is realized, various interferences are effectively inhibited, the signal characteristics are reserved, and the signal-to-noise ratio is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of signal acquisition and processing, and more particularly to a low-noise biopotential signal acquisition system and processing method. Background Art

[0002] Biopotential signals are electrical signals generated by human physiological activity, including electrocardiograms (ECGs), electroencephalograms (EEGs), and electromyograms (EMGs). They are a crucial source of information for clinical diagnosis, physiological monitoring, and human-computer interaction systems. These signals typically have extremely small amplitudes (microvolts to millivolts), a narrow frequency range, and are often obscured by various noises, posing significant challenges to their accurate acquisition and processing.

[0003] Traditional bioelectrical signal acquisition systems primarily consist of electrodes, amplifiers, filters, and data converters. Conventional surface electrodes form an interface with the skin through an electrolyte. This interface exhibits complex electrochemical properties, resulting in contact impedance and polarization effects, leading to signal attenuation and distortion. Furthermore, the impedance of the skin-electrode interface varies significantly over time, making long-term monitoring unstable.

[0004] Bioelectric signals themselves are weak (usually at the microvolt level) and are extremely susceptible to environmental electromagnetic interference, motion artifacts and physiological interference, resulting in unstable acquisition quality; in clinical practice, individual differences among patients are significant, and traditional fixed parameter systems cannot dynamically adapt to different skin impedances and changes in physiological states, often resulting in a significant decrease in signal quality during long-term monitoring; in environments with dense medical equipment, power frequency interference and electromagnetic coupling between devices seriously affect signal purity; in multi-channel acquisition, the problem of crosstalk between channels is prominent and the spatial resolution is limited; existing algorithms are not effective in processing non-stationary and nonlinear bioelectric signals, especially in motion or pathological states, where signal characteristics are complex and changeable, and traditional filtering methods often lose key physiological characteristics while removing noise; in addition, the lack of real-time quality assessment and feedback mechanisms makes it impossible for the system to make timely adjustments to sudden interference, and it is easy to miss important physiological information in key clinical scenarios such as emergency and intensive care, posing risks to medical decision-making.

[0005] In view of this, the present invention proposes a low-noise biopotential signal acquisition system and processing method to solve the above problems. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned objectives, the present invention provides the following technical solution: a low-noise biopotential signal acquisition and processing method, comprising:

[0007] Step 1: Collecting biopotential signals from the human body through an electrode array to obtain original biopotential signals;

[0008] Step 2: Input the original bioelectrical signal into a multi-level adaptive filtering and amplification circuit to obtain an amplified and filtered analog signal;

[0009] Step 3: Perform analog-to-digital conversion on the amplified and filtered analog signal to obtain a digitalized bioelectrical signal;

[0010] Step 4: Perform adaptive wavelet transform denoising on the digitalized bioelectrical signal to obtain a primary denoised signal;

[0011] Step 5: Perform empirical mode decomposition on the primary denoised signal to obtain a multi-level intrinsic mode function set;

[0012] Step 6: Extract the intrinsic mode functions of the characteristic frequency band from the multi-level intrinsic mode function set to obtain a characteristic intrinsic mode function set;

[0013] Step 7: Perform signal reconstruction based on the characteristic intrinsic mode function set to obtain a pure bioelectrical signal;

[0014] Step 8: Perform real-time quality assessment on the pure bioelectrical signal to obtain a signal quality index; dynamically adjust system parameters based on the signal quality index to optimize the bioelectrical signal acquisition process and obtain a high signal-to-noise ratio bioelectrical potential signal output.

[0015] Further, the obtaining of the original bioelectrical signal by collecting the bioelectrical potential signal from the human body through the electrode array includes:

[0016] Perform nano-material modification on the electrode surface to obtain a low-impedance bioelectrode, and arrange the low-impedance bioelectrodes to form an adaptive structure electrode array to obtain a multi-channel acquisition network;

[0017] Automatically measure the electrode-skin interface impedance of the multi-channel acquisition network to obtain interface contact quality data, and optimize channel selection according to the interface contact quality data to obtain an optimized acquisition channel group;

[0018] Real-time collect the changes in the bioelectrical potential on the human body surface through the optimized acquisition channel group to obtain multi-channel original signals, and perform inter-channel interference suppression on the multi-channel original signals to obtain a preliminary isolation signal;

[0019] Perform differential amplification preprocessing on the preliminary isolation signal to obtain a differential enhanced signal, and perform preliminary baseline drift correction on the differential enhanced signal to obtain a stable baseline signal;

[0020] Transmit the stable baseline signal to the subsequent processing circuit through a shielded transmission line to obtain the original bioelectrical signal.

[0021] Further, the step of inputting the original bioelectrical signal into a multi-stage adaptive filtering and amplifying circuit to obtain an amplified and filtered analog signal includes:

[0022] Input the original bioelectrical signal into a first-stage low-pass filter to obtain a first-stage filtered signal with high-frequency interference removed. The first-stage low-pass filter adopts a second-order Butterworth structure and has an adjustable cut-off frequency;

[0023] Input the first-stage filtered signal into a second-stage variable gain amplifier to obtain an amplified signal with optimized dynamic range. The gain parameter G of the second-stage variable gain amplifier is dynamically adjusted through real-time feedback control;

[0024] Input the amplified signal with optimized dynamic range into a third-stage dynamic band-pass filter to obtain a second-stage filtered signal with optimized frequency band. The center frequency fc and bandwidth parameter B of the third-stage dynamic band-pass filter are adaptively adjusted according to the signal characteristics;

[0025] Perform power frequency noise adaptive notch processing on the second-stage filtered signal with optimized frequency band to obtain a filtered signal with power frequency interference suppressed. The power frequency noise adaptive notch processing uses a phase-locked loop tracking technique to achieve compensation for power frequency drift;

[0026] Perform output buffering and impedance matching on the filtered signal with power frequency interference suppressed to obtain an amplified and filtered analog signal.

[0027] Further, the step of performing analog-to-digital conversion on the amplified and filtered analog signal to obtain a digitalized bioelectrical signal includes:

[0028] Perform signal conditioning on the amplified and filtered analog signal to obtain a standardized signal that meets the conversion range, and perform sample and hold on the standardized signal to obtain a discrete-time sampling point sequence;

[0029] Use a high-precision ADC to perform quantization encoding on the discrete-time sampling point sequence to obtain an initial digital code stream. The sampling rate of the high-precision ADC is dynamically adjusted according to the signal spectrum characteristics, and the resolution is not less than 16 bits;

[0030] Perform quantization noise shaping on the initial digital code stream to obtain an optimized digital signal, and perform digital filtering anti-aliasing processing on the optimized digital signal to obtain an anti-aliasing digital signal;

[0031] Perform timing calibration on the anti-aliasing digital signal to obtain a synchronized digital signal, and perform data format conversion on the synchronized digital signal to obtain digital data in a standard format;

[0032] Control the data stream of the digital data in the standard format through a buffer management mechanism to obtain a digitalized bioelectrical signal.

[0033] Further, the adaptive wavelet transform denoising process for the digitized bioelectrical signal to obtain a primary denoised signal includes:

[0034] Analyze the signal characteristics of the digitized bioelectrical signal to obtain a signal spectrum feature map, and select an optimal wavelet basis function according to the signal spectrum feature map to obtain an adapted wavelet basis;

[0035] Perform multi-scale wavelet decomposition on the digitized bioelectrical signal using the adapted wavelet basis to obtain a multi-layer wavelet coefficient set, and perform threshold analysis on the multi-layer wavelet coefficient set to obtain a noise threshold distribution;

[0036] Design an adaptive threshold function according to the noise threshold distribution to obtain an optimized threshold strategy, and apply the optimized threshold strategy to perform soft threshold processing on the wavelet coefficients to obtain denoised wavelet coefficients;

[0037] Apply a signal singularity detection algorithm to the denoised wavelet coefficients to obtain a singularity protection mapping, and perform coefficient correction according to the singularity protection mapping to obtain a wavelet coefficient set that preserves features;

[0038] Use the wavelet coefficient set that preserves features to perform inverse wavelet transform reconstruction to obtain a primary denoised signal.

[0039] Further, the empirical mode decomposition of the primary denoised signal to obtain a multi-level intrinsic mode function set includes:

[0040] Detect the extreme points of the primary denoised signal to obtain a signal extreme value distribution map, and construct upper and lower envelope lines based on the signal extreme value distribution map to obtain an envelope line pair;

[0041] Calculate the mean envelope line according to the envelope line pair to obtain a local mean function, and subtract the local mean function from the primary denoised signal to obtain a candidate intrinsic mode function;

[0042] Perform an intrinsicness test on the candidate intrinsic mode function to obtain an intrinsicness evaluation result. If the intrinsicness evaluation result does not meet the IMF condition, repeat the envelope calculation process until the IMF condition is met to obtain the first-layer intrinsic mode function;

[0043] Subtract the first-layer intrinsic mode function from the primary denoised signal to obtain a first residual signal, and repeat the empirical mode decomposition process on the first residual signal to obtain subsequent intrinsic mode functions;

[0044] Combine all the decomposed intrinsic mode functions and the final residual signal to obtain a complete multi-level intrinsic mode function set.

[0045] Further, extracting the intrinsic mode functions of the feature frequency bands from the multi-level intrinsic mode function set to obtain a feature intrinsic mode function set includes:

[0046] Performing Hilbert transform on each IMF in the multi-level intrinsic mode function set to obtain an instantaneous frequency distribution, and calculating the spectral center and bandwidth of each IMF according to the instantaneous frequency distribution to obtain an IMF spectral characteristic table;

[0047] Based on the IMF spectral characteristic table, performing bioelectric potential signal feature frequency band analysis to obtain a target frequency band range, and screening relevant IMFs according to the target frequency band range to obtain a preliminary selected IMF set;

[0048] Performing cross-correlation analysis on the preliminary selected IMF set to obtain an IMF correlation matrix, and removing redundant components according to the IMF correlation matrix to obtain a simplified IMF set;

[0049] Calculating the energy contribution degree of each IMF in the simplified IMF set to obtain an information quantity evaluation result, and performing importance ranking according to the information quantity evaluation result to obtain a ranked IMF set;

[0050] Based on the ranked IMF set, performing adaptive weight allocation to obtain a weighted IMF combination scheme, and constructing a feature intrinsic mode function set according to the weighted IMF combination scheme.

[0051] Further, reconstructing a signal based on the feature intrinsic mode function set to obtain a pure bioelectric signal includes:

[0052] Performing non-linear phase correction on the feature intrinsic mode function set to obtain a group of phase-aligned IMFs, and designing an optimal reconstruction weight vector for the group of phase-aligned IMFs to obtain an adaptive fusion strategy;

[0053] Based on the adaptive fusion strategy, performing weighted superposition on the IMFs to obtain a preliminary reconstructed signal, and performing morphological feature enhancement on the preliminary reconstructed signal to obtain a bioelectric signal with prominent features;

[0054] Performing physiological model constraint processing on the bioelectric signal with prominent features to obtain a signal that conforms to physiological characteristics, and performing final correction of baseline drift on the signal that conforms to physiological characteristics to obtain a stable zero-baseline signal;

[0055] Performing abnormal waveform detection and repair on the stable zero-baseline signal to obtain a corrected signal, and performing bandwidth optimization and smoothing processing on the corrected signal to obtain a final smooth signal;

[0056] Performing morphological matching verification on the final smooth signal with a standard bioelectric template to obtain a pure bioelectric signal.

[0057] Further, the real-time quality assessment of the pure bioelectric signal is performed to obtain a signal quality index, and based on the signal quality index, the system parameters are dynamically adjusted to optimize the bioelectric signal acquisition process, and a bioelectric potential signal output with a high signal-to-noise ratio is obtained, including:

[0058] Calculate the signal-to-noise ratio, effective bandwidth, and waveform distortion of the pure bioelectric signal to obtain a multi-dimensional quality assessment index, and construct a comprehensive quality function based on the multi-dimensional quality assessment index to obtain a signal quality index;

[0059] Establish a mapping relationship between the signal quality index and the key system parameters to obtain a parameter optimization model, and design a feedback control algorithm based on the parameter optimization model to obtain a system parameter adjustment strategy;

[0060] Based on the system parameter adjustment strategy, the gain parameter G of the second-stage variable gain amplifier is updated in real time to obtain dynamic optimized gain control, and the bandwidth parameter B of the third-stage dynamic band-pass filter is adaptively adjusted according to the signal characteristics to obtain an optimized frequency band setting;

[0061] Feed back the dynamic optimized gain control and the optimized frequency band setting to the hardware circuit to obtain a closed-loop self-optimizing structure, and based on the closed-loop self-optimizing structure, realize the dynamic adjustment of the electrode array channel configuration to obtain an optimized acquisition path;

[0062] Integrate the synergistic effects of the closed-loop self-optimizing structure and the optimized acquisition path to realize the adaptive optimization of the entire process of bioelectric potential signal acquisition, and obtain a bioelectric potential signal output with a high signal-to-noise ratio.

[0063] A low-noise bioelectric potential signal acquisition system, which is used to implement the low-noise bioelectric potential signal acquisition and processing method, includes: an electrode acquisition module, which is used to collect bioelectric potential signals from the human body through an electrode array to obtain an original bioelectric signal;

[0064] A multi-stage adaptive filtering and amplification module, which is used to input the original bioelectric signal into a multi-stage adaptive filtering and amplification circuit to obtain an amplified and filtered analog signal;

[0065] An analog-to-digital conversion module, which is used to perform analog-to-digital conversion on the amplified and filtered analog signal to obtain a digitized bioelectric signal;

[0066] A wavelet transform denoising module, which is used to perform adaptive wavelet transform denoising processing on the digitized bioelectric signal to obtain a primary denoised signal;

[0067] An empirical mode decomposition module, which is used to perform empirical mode decomposition on the primary denoised signal to obtain a multi-level intrinsic mode function set;

[0068] A feature extraction module, configured to extract the intrinsic mode functions of characteristic frequency bands from the multi-level intrinsic mode function set to obtain a characteristic intrinsic mode function set;

[0069] A signal reconstruction module, configured to reconstruct the signal based on the characteristic intrinsic mode function set to obtain a pure bioelectric signal;

[0070] The quality assessment and parameter optimization module is used to perform real-time quality assessment of the pure bioelectric signal to obtain a signal quality index; dynamically adjust system parameters based on the signal quality index, optimize the bioelectric signal acquisition process, and obtain a high signal-to-noise ratio biopotential signal output.

[0071] The technical effects and advantages of the low-noise biopotential signal acquisition system and processing method of the present invention are as follows:

[0072] The present invention improves the signal-to-noise ratio and fidelity of the signal, enabling the accurate capture of weak bioelectric signals in complex environments. It effectively overcomes the problems of signal instability, feature loss, and poor adaptability in traditional technologies, and achieves high-quality processing of non-stationary bioelectric signals. The closed-loop self-optimization mechanism established by the method can adjust system parameters in real time according to signal quality, giving the system excellent environmental adaptability and individual compatibility, providing a stable and reliable signal foundation for different application scenarios. By retaining the key physiological characteristics of the signal while effectively suppressing various types of interference, the accuracy and reliability of bioelectric signal analysis are significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 Schematic diagram of a low-noise biopotential signal acquisition and processing method of the present invention;

[0074] Figure 2 Detailed implementation flow chart of step 1 of the present invention;

[0075] Figure 3 Schematic diagram of a low-noise biopotential signal acquisition system of the present invention. DETAILED DESCRIPTION

[0076] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0077] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0078] The embodiments of the present application provide a low-noise bioelectric potential signal acquisition system and a processing method. The execution subjects of the system and the processing method include, but are not limited to, the following that carry the system: medical monitoring devices, bioelectric signal processing platforms, medical data analysis servers, wearable health monitoring devices, etc., which can be regarded as general computing nodes of the present application. The data processing platform includes, but is not limited to, at least one of a bioelectric signal acquisition system, a medical diagnosis information system, and an electrocardiogram monitoring data management system.

[0079] Please refer to Figure 1 , the present invention provides a low-noise bioelectric potential signal acquisition and processing method, including the following steps:

[0080] Step 1: Collect bioelectric potential signals from the human body through an electrode array to obtain original bioelectric signals;

[0081] Step 2: Input the original bioelectric signals into a multi-stage adaptive filter amplification circuit to obtain amplified and filtered analog signals;

[0082] Step 3: Perform analog-to-digital conversion on the amplified and filtered analog signals to obtain digitized bioelectric signals;

[0083] Step 4: Perform adaptive wavelet transform denoising processing on the digitized bioelectric signals to obtain primary denoised signals;

[0084] Step 5: Perform empirical mode decomposition on the primary denoised signals to obtain a multi-level intrinsic mode function set;

[0085] Step 6: Extract the intrinsic mode functions of the characteristic frequency bands from the multi-level intrinsic mode function set to obtain a characteristic intrinsic mode function set;

[0086] Step 7: Perform signal reconstruction based on the characteristic intrinsic mode function set to obtain pure bioelectric signals;

[0087] Step 8: Perform real-time quality assessment on the pure bioelectric signals to obtain signal quality indicators; dynamically adjust system parameters based on the signal quality indicators, optimize the bioelectric signal acquisition process, and obtain a high signal-to-noise ratio bioelectric potential signal output.

[0088] The present invention reduces the electrode-skin interface impedance and improves the signal acquisition quality by performing nano-material modification treatment on the electrode surface; adopts a multi-stage adaptive filter amplification circuit to effectively suppress various interferences; combines adaptive wavelet transform and empirical mode decomposition technologies to achieve efficient signal denoising and feature extraction; and dynamically optimizes system parameters through real-time quality assessment and closed-loop feedback control, and finally obtains a high signal-to-noise ratio bioelectric potential signal output.

[0089] In the embodiments of the present invention, the steps of the low-noise bioelectric potential signal acquisition and processing method include:

[0090] Step 1: Collect bioelectric potential signals from the human body through an electrode array to obtain the original bioelectric signals;

[0091] In this embodiment, first, the surfaces of the electrodes are modified with nanomaterials to obtain low-impedance bioelectrodes, and these electrodes are arranged into an electrode array with an adaptive structure to form a multi-channel acquisition network. Automatically measure the impedance of the electrode-skin interface to obtain interface contact quality data, and accordingly optimize the selection of acquisition channels. Real-time collect the bioelectric potential changes on the human body surface through the optimized acquisition channel group to obtain multi-channel original signals, and perform inter-channel interference suppression processing on these signals to obtain preliminarily isolated signals. Perform differential amplification preprocessing on the preliminarily isolated signals to enhance the useful signals, and then perform preliminary baseline drift correction to obtain stable baseline signals. Finally, transmit the signals to the subsequent processing circuit through a shielded transmission line to obtain the original bioelectric signals.

[0092] Step 2: Input the original bioelectric signals into a multi-stage adaptive filter amplification circuit to obtain amplified and filtered analog signals;

[0093] In this embodiment, the original bioelectric signals are first input into a first-stage low-pass filter, which adopts a second-order Butterworth structure with an adjustable cut-off frequency and is used to remove high-frequency interference to obtain a first-stage filtered signal. Then, the first-stage filtered signal is input into a second-stage variable gain amplifier, and its gain parameter G is dynamically adjusted through real-time feedback control to optimize the dynamic range of the signal to obtain an amplified signal with an optimized dynamic range. Next, the amplified signal is input into a third-stage dynamic band-pass filter, and the center frequency fc and bandwidth parameter B of this filter are adaptively adjusted according to the signal characteristics to obtain a second-stage filtered signal with an optimized frequency band. Perform adaptive notch processing on the second-stage filtered signal for power frequency noise, and use a phase-locked loop tracking technique to achieve compensation for power frequency drift and effectively suppress power frequency interference. Finally, perform output buffering and impedance matching on the filtered signal with power frequency interference suppressed to obtain amplified and filtered analog signals.

[0094] Step 3: Perform analog-to-digital conversion on the amplified and filtered analog signals to obtain digitalized bioelectric signals;

[0095] In this embodiment, first, the amplified and filtered analog signal is subjected to signal conditioning to meet the conversion range requirements, obtaining a standardized signal, and the standardized signal is sampled and held to obtain a discrete-time sampling point sequence. Then, a high-precision ADC is used to quantize and encode the discrete-time sampling point sequence to obtain an initial digital code stream. The sampling rate of this ADC is dynamically adjusted according to the signal spectrum characteristics, and the resolution is not less than 16 bits. The initial digital code stream is subjected to quantization noise shaping to obtain an optimized digital signal, and digital filtering anti-aliasing processing is performed to obtain an anti-aliasing digital signal. Then, the anti-aliasing digital signal is subjected to timing calibration to obtain a synchronous digital signal, and data format conversion is performed to obtain digital data in a standard format. Finally, data stream control is performed through a buffer management mechanism to obtain a digitized bioelectric signal.

[0096] Step 4: Perform adaptive wavelet transform denoising processing on the digitized bioelectric signal to obtain a primary denoised signal;

[0097] In this embodiment, first, the signal characteristics of the digitized bioelectric signal are analyzed to obtain a signal spectrum feature map, and based on this, the optimal wavelet basis function is selected to obtain an adapted wavelet basis. The digitized bioelectric signal is subjected to multi-scale wavelet decomposition using the adapted wavelet basis to obtain a multi-layer wavelet coefficient set, and threshold analysis is performed on these coefficients to obtain a noise threshold distribution. An adaptive threshold function is designed according to the noise threshold distribution to obtain an optimized threshold strategy, and this strategy is applied to perform soft threshold processing on the wavelet coefficients to obtain denoised wavelet coefficients. The signal singularity detection algorithm is applied to the denoised wavelet coefficients to obtain a singularity protection mapping, and based on this, coefficient correction is performed to obtain a wavelet coefficient set that preserves features. Finally, inverse wavelet transform reconstruction is performed using the wavelet coefficient set that preserves features to obtain a primary denoised signal.

[0098] Step 5: Perform empirical mode decomposition on the primary denoised signal to obtain a multi-level intrinsic mode function set;

[0099] In this embodiment, first, the extreme points of the primary denoised signal are detected to obtain a signal extreme value distribution map, and based on this distribution map, upper and lower envelope lines are constructed to obtain an envelope line pair. The mean envelope line is calculated according to the envelope line pair to obtain a local mean function, and the primary denoised signal is subtracted from the local mean function to obtain a candidate intrinsic mode function. Intrinsicness testing is performed on the candidate intrinsic mode function to obtain an intrinsicness evaluation result. If the evaluation result does not meet the IMF conditions, the envelope calculation process is repeated until the IMF conditions are met to obtain the first-layer intrinsic mode function. The primary denoised signal is subtracted from the first-layer intrinsic mode function to obtain a first residual signal, and the empirical mode decomposition process is repeatedly performed on the first residual signal to obtain subsequent intrinsic mode functions. Finally, all the decomposed intrinsic mode functions and the final residual signal are combined to obtain a complete multi-level intrinsic mode function set.

[0100] Step 6: Extract the intrinsic mode functions in the characteristic frequency band from the multi-level intrinsic mode function set to obtain a set of characteristic intrinsic mode functions;

[0101] In this embodiment, first, perform Hilbert transform on each IMF in the multi-level intrinsic mode function set to obtain the instantaneous frequency distribution, and calculate the spectral center and bandwidth of each IMF according to the instantaneous frequency distribution to obtain the IMF spectral characteristic table. Based on the IMF spectral characteristic table, perform bioelectrical signal characteristic frequency band analysis to obtain the target frequency band range, and screen relevant IMFs according to the target frequency band range to obtain a preliminary IMF set. Perform cross-correlation analysis on the preliminary IMF set to obtain the IMF correlation matrix, and remove redundant components according to the correlation matrix to obtain a simplified IMF set. Calculate the energy contribution degree of each IMF in the simplified IMF set to obtain the information quantity evaluation result, and perform importance ranking according to the evaluation result to obtain a ranked IMF set. Based on the ranked IMF set, perform adaptive weight allocation to obtain a weighted IMF combination scheme, and construct a set of characteristic intrinsic mode functions according to the combination scheme.

[0102] Step 7: Based on the set of characteristic intrinsic mode functions, perform signal reconstruction to obtain a pure bioelectrical signal;

[0103] In this embodiment, first, perform non-linear phase correction on the set of characteristic intrinsic mode functions to obtain a group of IMFs with phase alignment, and design an optimal reconstruction weight vector for the group of IMFs with phase alignment to obtain an adaptive fusion strategy. Based on the adaptive fusion strategy, perform weighted superposition on the IMFs to obtain a preliminary reconstructed signal, and perform morphological feature enhancement on the preliminary reconstructed signal to obtain a bioelectrical signal with prominent features. Perform physiological model constraint processing on the bioelectrical signal with prominent features to obtain a signal that conforms to physiological characteristics, and perform final correction of baseline drift on the signal that conforms to physiological characteristics to obtain a stable zero baseline signal. Perform abnormal waveform detection and repair on the stable zero baseline signal to obtain a corrected signal, and perform bandwidth optimization and smoothing processing on the corrected signal to obtain a final smooth signal. Perform morphological matching verification on the final smooth signal and the standard bioelectrical template to obtain a pure bioelectrical signal.

[0104] Step 8: Perform real-time quality assessment on the pure bioelectrical signal to obtain a signal quality index; Dynamically adjust the system parameters based on the signal quality index to optimize the bioelectrical signal acquisition process and obtain a bioelectrical signal output with a high signal-to-noise ratio.

[0105] In this embodiment, the signal-to-noise ratio, effective bandwidth and waveform distortion of the pure bioelectric signal are first calculated to obtain a multidimensional quality evaluation index, and a comprehensive quality function is constructed based on the multidimensional quality evaluation index to obtain a signal quality index. A mapping relationship between the signal quality index and the key system parameters is established to obtain a parameter optimization model, and a feedback control algorithm is designed based on the parameter optimization model to obtain a system parameter adjustment strategy. Based on the system parameter adjustment strategy, the gain parameter G of the second-stage variable gain amplifier is updated in real time to obtain a dynamic optimized gain control, and the bandwidth parameter B of the third-stage dynamic bandpass filter is adaptively adjusted according to the signal characteristics to obtain an optimized frequency band setting. The dynamic optimized gain control and the optimized frequency band setting are fed back to the hardware circuit to obtain a closed-loop self-optimization structure, and the electrode array channel configuration is dynamically adjusted based on the closed-loop self-optimization structure to obtain an optimized acquisition path. The synergistic effect of the comprehensive closed-loop self-optimization structure and the optimized acquisition path is used to achieve adaptive optimization of the entire biopotential signal acquisition process and obtain a high signal-to-noise ratio biopotential signal output.

[0106] See Figure 2 , is a flowchart of the detailed implementation steps of step 1. In this embodiment, the detailed implementation steps of step 1 include:

[0107] The electrode surface is modified with nanomaterials to obtain a low-impedance bioelectrode, and the low-impedance bioelectrode is arranged to form an adaptive structure electrode array to obtain a multi-channel acquisition network;

[0108] Automatically measuring the electrode-skin interface impedance of the multi-channel acquisition network to obtain interface contact quality data, and optimizing channel selection based on the interface contact quality data to obtain an optimized acquisition channel group;

[0109] The optimized acquisition channel group is used to collect changes in human body surface biopotential in real time to obtain multi-channel original signals, and inter-channel interference suppression is performed on the multi-channel original signals to obtain preliminary isolated signals;

[0110] Performing differential amplification pre-processing on the preliminary isolated signal to obtain a differential enhanced signal, and performing preliminary baseline drift correction on the differential enhanced signal to obtain a stable baseline signal;

[0111] The stable baseline signal is transmitted to a subsequent processing circuit through a shielded transmission line to obtain an original bioelectric signal.

[0112] In this embodiment, the surface of the traditional bioelectrode is modified with nanomaterials (such as silver nanoparticles, gold nanoparticles, graphene, etc.), significantly reducing the impedance at the electrode-skin contact interface and improving the signal acquisition quality. The nanomaterials are fixed on the electrode surface through physical deposition or chemical modification methods to form a porous nanostructure, increasing the effective surface area of the electrode, while improving conductivity and biocompatibility. The low-impedance bioelectrodes are arranged into an electrode array with an adaptive structure according to the human anatomical characteristics and the characteristics of the target bioelectric signals to form a multi-channel acquisition network, improving the spatial resolution and anti-interference ability.

[0113] Using the built-in impedance measurement circuit, the impedance of the electrode-skin interface is measured automatically in real time to obtain the interface contact quality data. The measurement principle is based on injecting a weak AC signal (usually 1 - 10 kHz, <10 μA), and analyzing the voltage response to calculate the impedance value. According to the interface contact quality data, an optimization algorithm (such as greedy search or genetic algorithm) is used to optimize the channel selection, screening out the electrode combination with the best signal quality to form an optimized acquisition channel group.

[0114] The bioelectric potential changes on the human body surface are collected in real time through the optimized acquisition channel group to obtain multi-channel raw signals. During the acquisition process, the driven right leg circuit (DRL) technology is applied to suppress the common-mode interference and improve the signal quality. The spatial filtering algorithm (such as Laplacian filtering or independent component analysis) is applied to the multi-channel raw signals to effectively suppress the crosstalk and interference between channels and obtain a preliminary isolated signal.

[0115] The preliminary isolated signal is pre-processed by differential amplification using an instrumentation amplifier with a high input impedance (>10 GΩ) and low noise (<1 μVrms) to enhance the useful signal and further suppress the common-mode interference at the same time, obtaining a differentially enhanced signal. The differentially enhanced signal is subjected to high-pass filtering (cutoff frequency 0.05 - 0.1 Hz) for preliminary correction of the baseline drift to eliminate the low-frequency drift caused by factors such as breathing and body movement, obtaining a stable baseline signal.

[0116] The stable baseline signal is transmitted to the subsequent processing circuit through a multi-layer shielded transmission line, and the shield layer is grounded to effectively isolate external electromagnetic interference. The line uses a differential transmission method to further improve the anti-interference ability. The signal integrity is maintained during the signal transmission process, and finally the original bioelectric signal is obtained.

[0117] In this embodiment, the detailed implementation steps of step 2 include:

[0118] Input the original bioelectric signal into the first-stage low-pass filter to obtain a first-stage filtered signal with high-frequency interference removed. The first-stage low-pass filter adopts a second-order Butterworth structure and the cut-off frequency is adjustable;

[0119] Input the first-stage filtered signal into a second-stage variable gain amplifier to obtain an amplified signal with optimized dynamic range. The gain parameter G of the second-stage variable gain amplifier is dynamically adjusted through real-time feedback control;

[0120] Input the amplified signal with optimized dynamic range into a third-stage dynamic band-pass filter to obtain a second-stage filtered signal with optimized frequency band. The center frequency fc and bandwidth parameter B of the third-stage dynamic band-pass filter are adaptively adjusted according to the signal characteristics;

[0121] Perform power frequency noise adaptive notch filtering on the second-stage filtered signal with optimized frequency band to obtain a filtered signal with power frequency interference suppressed. The power frequency noise adaptive notch filtering is implemented using a phase-locked loop tracking technique to compensate for power frequency drift;

[0122] Perform output buffering and impedance matching on the filtered signal with power frequency interference suppressed to obtain an amplified and filtered analog signal.

[0123] In this embodiment, the original bioelectric signal is input into a first-stage low-pass filter. This filter adopts a second-order Butterworth structure, has a flat passband response and a relatively steep transition band, and effectively suppresses high-frequency interference. The cut-off frequency of the filter can be adjusted within the range of 1 - 500 Hz to adapt to the frequency characteristics of different types of bioelectric signals (such as electrocardiogram, electromyogram, electroencephalogram, etc.). The low-pass filter is implemented using a precision operational amplifier and high-precision passive components to ensure the stability and consistency of the filtering performance, and obtain a first-stage filtered signal with high-frequency interference removed.

[0124] Input the first-stage filtered signal into a second-stage variable gain amplifier. This amplifier is constructed based on a precision instrumentation amplifier, has a high common-mode rejection ratio (>100 dB) and low noise characteristics (<0.5 μVrms). The gain parameter G of the amplifier can be continuously adjusted within the range of 10 - 10000, and is realized through a digital potentiometer or a digital gain amplifier. The gain parameter G is dynamically adjusted through a real-time feedback control system, automatically optimizes the gain setting according to the signal amplitude, and avoids problems such as signal truncation or insufficient utilization of the dynamic range, and obtains an amplified signal with optimized dynamic range.

[0125] Input the amplified signal with optimized dynamic range into a third-stage dynamic band-pass filter. This filter adopts a state variable filter structure and has independently adjustable center frequency fc and bandwidth parameter B. The center frequency fc can be adjusted within the range of 0.5 - 200 Hz, and the bandwidth parameter B can be adjusted within the range of 0.1 - 50 Hz. The best parameters are adaptively set according to the spectral characteristics of the target bioelectric signal. The filter is designed with a high quality factor Q to ensure that the in-band signal passes through while effectively suppressing out-of-band interference, and obtains a second-stage filtered signal with optimized frequency band.

[0126] Perform adaptive notch filtering on the second-stage filtered signal optimized for the frequency band to effectively suppress 50 / 60 Hz power frequency interference and its harmonics. The notch filter uses a phase-locked loop (PLL) tracking technique to detect the actual frequency of the power frequency signal in real time (considering that the power grid frequency has a drift of about ±0.5 Hz), and dynamically adjusts the center frequency of the notch filter accordingly to ensure that the notch filter is always aligned with the actual power frequency to maximize the suppression effect. The notch filter adopts a high-Q narrowband design to minimize the impact on the useful signal and obtain a filtered signal that suppresses power frequency interference.

[0127] Perform output buffering and impedance matching on the filtered signal that suppresses power frequency interference. Use a buffer amplifier with a low output impedance (<50 Ω) to ensure that the signal is not affected by the load during transmission. At the same time, perform impedance matching to match the output impedance with the input impedance of the subsequent analog-to-digital conversion circuit to maximize the signal transmission efficiency and reduce reflection and distortion. The buffer circuit has sufficient bandwidth and driving ability to ensure the complete transmission of the signal and obtain an amplified and filtered analog signal.

[0128] In this embodiment, the detailed implementation steps of step 3 include:

[0129] Perform signal conditioning on the amplified and filtered analog signal to obtain a standardized signal that meets the conversion range, and perform sample and hold on the standardized signal to obtain a discrete-time sampling point sequence;

[0130] Use a high-precision ADC to perform quantization encoding on the discrete-time sampling point sequence to obtain an initial digital code stream. The sampling rate of the high-precision ADC is dynamically adjusted according to the signal spectrum characteristics, and the resolution is not less than 16 bits;

[0131] Perform quantization noise shaping on the initial digital code stream to obtain an optimized digital signal, and perform digital filtering anti-aliasing processing on the optimized digital signal to obtain an anti-aliasing digital signal;

[0132] Perform timing calibration on the anti-aliasing digital signal to obtain a synchronous digital signal, and perform data format conversion on the synchronous digital signal to obtain a standard format digital data;

[0133] Control the data stream of the standard format digital data through a buffer management mechanism to obtain a digitized bioelectrical signal.

[0134] In this embodiment, the amplified and filtered analog signal is signal-conditioned to ensure that its amplitude range matches the input range of the ADC (such as ±2.5V), avoiding truncation and low quantization efficiency. The conditioning circuit includes a precision amplifier and a level biasing circuit, which can perform automatic gain control and zero calibration according to the signal characteristics to obtain a standardized signal that meets the conversion range. The standardized signal is sampled using a high-performance sample-and-hold circuit, which has a fast acquisition time (<100ns) and a low voltage drop characteristic (<0.01%), ensuring sampling accuracy. The sampling frequency is dynamically adjusted according to the signal bandwidth, following the Nyquist sampling theorem, and is generally set to 2.5 - 5 times the highest frequency of the signal to obtain a discrete-time sampling point sequence.

[0135] A high-precision ADC is used to quantize and encode the discrete-time sampling point sequence, converting the analog signal into a digital representation. The ADC uses a 16 - 24-bit high-resolution architecture (such as successive approximation or Sigma-Delta type) to ensure that small changes in the bioelectric signal can be accurately captured. The sampling rate of the ADC is dynamically adjusted according to the signal spectral characteristics, usually in the range of 1kHz - 10kHz, to ensure sufficient time resolution. The ADC has characteristics of low noise (<1μVrms), high linearity (INL < 0.5LSB), and low power consumption (<10mW), making it suitable for portable bioelectric signal acquisition devices. The quantization process generates an initial digital code stream, representing the amplitude of the sampled signal in binary form.

[0136] Digital signal processing techniques are applied to the initial digital code stream for quantization noise shaping, redistributing the quantization noise power spectrum, pushing the noise out of the signal band, and improving the signal-to-noise ratio within the signal band. Common techniques include oversampling and Sigma-Delta modulation, which can increase the effective number of bits by 1 - 4 bits. After quantization noise shaping, an optimized digital signal is obtained, and then a digital filter is applied for anti-aliasing processing to filter out the aliasing components that may be introduced during the sampling process. The digital filter uses an FIR structure and has a linear phase characteristic to ensure that the signal waveform is not distorted, obtaining an anti-aliasing digital signal.

[0137] Timing calibration is performed on the anti-aliasing digital signal to compensate for the time delays introduced by each link in the system, ensuring the accurate phase relationship between multi-channel signals. The calibration process uses correlation analysis and interpolation techniques to precisely align the signal timing, obtaining a synchronized digital signal. Data format conversion is performed on the synchronized digital signal, converting the original binary data into a standardized format (such as IEEE754 floating-point format or fixed-point format) for subsequent processing and storage. Unit conversion is also performed during the format conversion, converting the digital code output by the ADC into the actual physical quantity unit (such as μV) to obtain the digital data in standard format.

[0138] The standard - format digital data is subjected to data - flow control through a buffer - management mechanism to balance the difference between the data - generation rate and the processing rate, preventing data loss or processing delay. The buffer adopts a FIFO (First - In - First - Out) structure and has overflow - protection and underflow - handling mechanisms to ensure the continuity and integrity of the data flow. The buffer size is dynamically adjusted according to the system processing capacity and the data - generation rate, usually ranging from 1 to 32 KB, and finally a stable and reliable digital bio - electrical signal is obtained.

[0139] In this embodiment, the detailed implementation steps of step 4 include:

[0140] Perform signal - characteristic analysis on the digital bio - electrical signal to obtain a signal - spectrum feature map, and select an optimal wavelet - basis function according to the signal - spectrum feature map to obtain an adapted wavelet basis;

[0141] Perform multi - scale wavelet decomposition on the digital bio - electrical signal using the adapted wavelet basis to obtain a multi - layer wavelet - coefficient set, and perform threshold analysis on the multi - layer wavelet - coefficient set to obtain a noise - threshold distribution;

[0142] Design an adaptive threshold function according to the noise - threshold distribution to obtain an optimized threshold strategy, and apply the optimized threshold strategy to perform soft - threshold processing on the wavelet coefficients to obtain denoised wavelet coefficients;

[0143] Apply a signal - singularity detection algorithm to the denoised wavelet coefficients to obtain a singularity - protection mapping, and perform coefficient correction according to the singularity - protection mapping to obtain a wavelet - coefficient set that preserves features;

[0144] Perform inverse wavelet transform reconstruction using the wavelet - coefficient set that preserves features to obtain a primary denoised signal.

[0145] In this embodiment, first, time - frequency analysis is performed on the digital bio - electrical signal, including short - time Fourier transform (STFT) and preliminary wavelet transform analysis, to obtain the time - frequency characteristics of the signal. Calculate characteristic parameters such as the energy distribution, spectral center, bandwidth, and peak frequency of the signal, and construct a signal - spectrum feature map. According to the signal - spectrum feature map, select an optimal wavelet - basis function from a candidate wavelet library (including wavelet families such as Daubechies, Symlet, Coiflet, Biorthogonal, etc.). The selection criteria are based on indicators such as the similarity between the wavelet basis and the signal morphology, energy concentration, and entropy minimization, and finally determine the adapted wavelet basis that is most suitable for the characteristics of the current bio - electrical signal.

[0146] The digital bioelectrical signal is subjected to multi-scale wavelet decomposition using an adaptive wavelet basis, and the signal is decomposed into wavelet coefficients at different frequency scales. The number of decomposition layers is determined according to the spectral characteristics of the signal, usually 3 - 8 layers, to ensure that the main frequency components of the signal can be fully separated. The decomposition process is implemented using the fast wavelet transform algorithm to obtain a multi-layer wavelet coefficient set containing approximation coefficients and detail coefficients. Statistical analysis is performed on the detail coefficients of each layer of the multi-layer wavelet coefficient set to estimate the distribution characteristics of the noise components. The noise level is estimated based on the MAD (Median Absolute Deviation) method, and the noise threshold at each scale is calculated to obtain the noise threshold distribution.

[0147] According to the noise threshold distribution, an adaptive threshold function is designed. This function can dynamically adjust the threshold size according to different decomposition scales and the local characteristics of the signal. The threshold function adopts a scale-dependent design. A smaller threshold is used at low-frequency scales to retain the signal structure, and a larger threshold is used at high-frequency scales to enhance the noise reduction effect. At the same time, a local signal energy factor is introduced to reduce the threshold in the region where the signal energy is concentrated to protect the useful information, obtaining an optimized threshold strategy. The optimized threshold strategy is used to perform soft threshold processing on the wavelet coefficients. The soft threshold function is defined as:

[0148] η(d, λ) = sign(d)·max(0, |d| - λ); where d is the wavelet coefficient and λ is the threshold. Soft threshold processing can effectively reduce the discontinuity of the processed signal compared to hard thresholding, obtaining denoised wavelet coefficients.

[0149] The signal singularity detection algorithm is applied to the denoised wavelet coefficients to identify important feature points in the signal (such as key waveforms like spikes and steep changes in bioelectrical signals). The detection algorithm is based on Lipschitz exponent analysis, calculates the attenuation characteristics of wavelet coefficients at different scales, and determines whether the coefficients correspond to signal singularity points. According to the detection results, a singularity protection mapping is constructed, which marks the positions of wavelet coefficients that need special protection. The denoised wavelet coefficients are corrected according to the singularity protection mapping. The coefficients corresponding to the singularity points are reduced or the threshold processing is cancelled to ensure that important features are not over-smoothed, obtaining a wavelet coefficient set that preserves features.

[0150] The inverse wavelet transform is performed using the wavelet coefficient set that preserves features to reconstruct the time-domain signal. The same wavelet basis function as that used in the decomposition is adopted during the reconstruction process to ensure the integrity and consistency of the signal. The reconstruction algorithm is implemented using the fast inverse wavelet transform, which has high computational efficiency and is suitable for real-time processing requirements. Boundary processing techniques are applied during the reconstruction process to reduce the influence of boundary effects on the signal. The reconstructed signal retains the main features of the original bioelectrical signal and effectively suppresses the noise components, forming a primary denoised signal.

[0151] In this embodiment, the detailed implementation steps of step 5 include:

[0152] Perform extreme point detection on the primary noise-reduced signal to obtain a signal extreme value distribution diagram, and construct upper and lower envelope lines based on the signal extreme value distribution diagram to obtain an envelope line pair;

[0153] Calculate the mean envelope line according to the envelope line pair to obtain a local mean function, and subtract the local mean function from the primary noise-reduced signal to obtain a candidate intrinsic mode function;

[0154] Perform an intrinsic property test on the candidate intrinsic mode function to obtain an intrinsic property evaluation result. If the intrinsic property evaluation result does not meet the IMF condition, repeat the envelope calculation process until the IMF condition is met to obtain the first-layer intrinsic mode function;

[0155] Subtract the first-layer intrinsic mode function from the primary noise-reduced signal to obtain a first residual signal, and repeat the empirical mode decomposition process on the first residual signal to obtain subsequent intrinsic mode functions;

[0156] Combine all the decomposed intrinsic mode functions and the final residual signal to obtain a complete multi-level intrinsic mode function set.

[0157] In this embodiment, first, perform extreme point detection on the primary noise-reduced signal to identify all local maximum points and minimum points in the signal. The detection algorithm is based on derivative analysis and uses the three-point comparison method to identify extreme points. At the same time, smooth preprocessing is applied to reduce the influence of false extreme points. Arrange the detected extreme points in chronological order to form a signal extreme value distribution diagram, which reflects the oscillation characteristics of the signal. Based on the signal extreme value distribution diagram, construct the upper envelope line (connecting all maximum points) and the lower envelope line (connecting all minimum points) of the signal through interpolation methods respectively. The interpolation method uses cubic spline interpolation to ensure the smoothness and accuracy of the envelope line, and obtains an envelope line pair.

[0158] Calculate the arithmetic mean of the upper and lower envelope lines to obtain the mean envelope line, and this curve represents the local mean function of the signal. The local mean function reflects the low-frequency trend component of the signal. Subtract the local mean function from the primary noise-reduced signal to obtain a candidate intrinsic mode function. The candidate intrinsic mode function should theoretically be a zero-mean oscillating function, but in fact, it may need further processing to meet the strict definition of IMF (intrinsic mode function).

[0159] Perform an intrinsic property test on the candidate intrinsic mode function to evaluate whether it meets the two conditions of IMF:

[0160] 1. The difference between the number of extreme points and the number of zero-crossing points does not exceed 1;

[0161] 2. The local mean function approaches zero. Calculate the standard deviation ratio SD of the candidate IMF:

[0162] Among them, h(k - 1)(t) and hk(t) are the candidate IMFs of the previous and current iterations respectively. When the SD is less than the preset threshold (usually 0.2 - 0.3), it is considered that the candidate IMF meets the intrinsic condition. If the result of the intrinsic evaluation does not meet the IMF condition, the candidate IMF is used as the new input signal, and the processes of envelope construction and mean function calculation are repeatedly executed for screening and iteration until the IMF condition is met or the maximum number of iterations (usually 10 times) is reached, and the first-layer intrinsic mode function is obtained.

[0163] Subtract the first-layer intrinsic mode function from the primary denoised signal to obtain the first residual signal. The first residual signal contains the remaining low-frequency components in the signal. Repeat the empirical mode decomposition process for it to obtain the second-layer intrinsic mode function. Continue this process to extract the intrinsic mode functions layer by layer until the residual signal becomes a monotonic function or its energy is lower than the preset threshold and no meaningful IMF can be extracted anymore. Combine all the decomposed intrinsic mode functions (arranged from high frequency to low frequency) and the final residual signal (representing the signal trend) together to form a complete multi-level intrinsic mode function set, which completely expresses the oscillation modes of the original signal at different characteristic scales.

[0164] In this embodiment, the detailed implementation steps of step 6 include:

[0165] Perform Hilbert transform on each IMF in the multi-level intrinsic mode function set to obtain the instantaneous frequency distribution, and calculate the spectral center and bandwidth of each IMF according to the instantaneous frequency distribution to obtain the IMF spectral characteristic table;

[0166] Based on the IMF spectral characteristic table, perform bioelectric potential signal characteristic frequency band analysis to obtain the target frequency band range, and screen relevant IMFs according to the target frequency band range to obtain the primary selected IMF set;

[0167] Perform cross-correlation analysis on the primary selected IMF set to obtain the IMF correlation matrix, and remove redundant components according to the IMF correlation matrix to obtain the simplified IMF set;

[0168] Calculate the energy contribution degree of each IMF in the simplified IMF set to obtain the information quantity evaluation result, and perform importance ranking according to the information quantity evaluation result to obtain the ranked IMF set;

[0169] Based on the ranked IMF set, perform adaptive weight allocation to obtain the weighted IMF combination scheme, and construct the characteristic intrinsic mode function set according to the weighted IMF combination scheme.

[0170] In this embodiment, first, the Hilbert transform is applied to each IMF in the multi-level intrinsic mode function set to convert the real signal into an analytic signal, obtaining the instantaneous amplitude and instantaneous phase of the signal. The instantaneous frequency distribution is obtained by calculating the derivative of the instantaneous phase. According to the instantaneous frequency distribution, the spectral center (average instantaneous frequency) and bandwidth (standard deviation of the instantaneous frequency) of each IMF are calculated, and an IMF spectral characteristic table is constructed, which comprehensively describes the frequency characteristics of each IMF.

[0171] Based on the IMF spectral characteristic table, combined with the known frequency characteristics of the target bioelectrical potential signal (for example, the main frequency of the electrocardiogram signal is 0.5 - 40 Hz, the main frequency of the electroencephalogram signal is 0.5 - 30 Hz, and the main frequency of the electromyogram signal is 20 - 500 Hz), characteristic frequency band analysis is carried out to determine the target frequency band range containing useful information. According to the target frequency band range, the IMFs whose spectral centers fall within this range are screened to form a primary selected IMF set. The screening process considers the frequency overlap degree to ensure the complete coverage of important frequency bands, and at the same time excludes the IMF components that significantly exceed the target range.

[0172] Perform cross-correlation analysis on each pair of IMFs (IMF_i and IMF_j) in the primary selected IMF set, and calculate their cross-correlation coefficient ρ_ij:

[0173] where μ_i and μ_j are the means of IMF_i and IMF_j respectively, and σ_i and σ_j are their standard deviations respectively. Organize all cross-correlation coefficients into an n×n matrix (n is the number of primary selected IMFs) to obtain the IMF correlation matrix. Identify the highly correlated IMF pairs (absolute value of the correlation coefficient > 0.7) according to the correlation matrix, and retain the one with a large amount of information in each pair of highly correlated IMFs by methods such as principal component analysis or information entropy comparison, and remove the redundant components to obtain a simplified IMF set.

[0174] Calculate the energy contribution degree of each IMF (IMF_i) in the simplified IMF set to evaluate the effective information amount it contains. The calculation formula for the energy contribution degree is:

[0175] where x(t) is the original signal. At the same time, combined with indicators such as the correlation degree and signal-to-noise ratio between the IMF and the target bioelectrical potential signal template, comprehensively evaluate the information value of each IMF to obtain the information amount evaluation result. Sort the IMFs in the simplified IMF set according to the information amount evaluation result, and arrange them in descending order of information contribution to obtain a sorted IMF set.

[0176] Based on the sorted IMF set, an adaptive weight allocation strategy is designed to assign reconstruction weights to different IMFs. The weight allocation takes into account factors such as the energy contribution of the IMF, the matching degree of the frequency characteristics with the target signal, and the signal-to-noise ratio. An optimization algorithm (such as a genetic algorithm or gradient descent) is used to solve for the optimal weight combination. After obtaining the weighted IMF combination scheme, the key components in the sorted IMF set are selected and integrated according to this scheme to construct a characteristic intrinsic mode function set, which contains the main characteristic components of the bioelectric potential signal and provides a basis for subsequent signal reconstruction.

[0177] In this embodiment, the detailed implementation steps of step 7 include:

[0178] Perform non-linear phase correction on the characteristic intrinsic mode function set to obtain a phase-aligned IMF group, and design an optimal reconstruction weight vector for the phase-aligned IMF group to obtain an adaptive fusion strategy;

[0179] Based on the adaptive fusion strategy, perform weighted superposition on the IMFs to obtain a preliminary reconstructed signal, and perform morphological feature enhancement on the preliminary reconstructed signal to obtain a bioelectric signal with prominent features;

[0180] Perform physiological model constraint processing on the bioelectric signal with prominent features to obtain a signal that conforms to physiological characteristics, and perform final baseline drift correction on the signal that conforms to physiological characteristics to obtain a stable zero-baseline signal;

[0181] Perform abnormal waveform detection and repair on the stable zero-baseline signal to obtain a corrected signal, and perform bandwidth optimization and smoothing processing on the corrected signal to obtain a final smooth signal;

[0182] Perform morphological matching verification on the final smooth signal with a standard bioelectric template to obtain a pure bioelectric signal.

[0183] In this embodiment, first, non-linear phase correction is performed on each IMF in the characteristic intrinsic mode function set to solve the problems of modal aliasing and phase distortion that may be introduced during the EMD decomposition process. The phase correction uses group delay compensation technology. According to the frequency characteristics of each IMF, a phase correction filter is designed to adjust the phase relationship between the IMFs so that they can be correctly aligned during reconstruction, obtaining a phase-aligned IMF group. An optimal reconstruction weight vector is designed for the phase-aligned IMF group, and this vector determines the contribution ratio of each IMF during the reconstruction process. The weight design uses an objective function optimization method to minimize the error between the reconstructed signal and the ideal bioelectric signal template, while considering smoothness and physiological rationality constraints, obtaining an adaptive fusion strategy.

[0184] Based on the adaptive fusion strategy, perform weighted superposition on the IMFs. The reconstruction formula is:

[0185] $x_1(t)=\sum_{i}[w_i\cdot IMF_i(t)] + rn(t)$; where $w_i$ is the IMF weight and $rn(t)$ is the residual signal. In the weighted superposition process, window function technology is applied to handle boundary effects, ensuring the stability of the reconstructed signal within the entire time domain to obtain a preliminary reconstructed signal. The morphological algorithm is applied to the preliminary reconstructed signal for feature enhancement to highlight the key waveform features of the bioelectric signal (such as the QRS complex of the electrocardiogram signal, the alpha wave of the electroencephalogram signal, etc.). The enhancement process uses an adaptive morphological operator to dynamically adjust the structural element according to the signal characteristics, retaining the features while avoiding introducing artifacts, resulting in a bioelectric signal with prominent features.

[0186] The bioelectric signal with prominent features is processed using physiological model constraints to ensure that the signal conforms to the characteristics of bioelectric physiology. The constraint processing is based on bioelectric signal models constructed from prior knowledge (such as the Pan-Tompkins model for electrocardiogram signals, the ARMA model for electroencephalogram signals, etc.), and rationality tests and adjustments are performed on parameters such as the waveform, amplitude range, and time relationship of the signal. The constraint processing uses a Bayesian inference framework to correct abnormal components that do not conform to physiological laws while maintaining the main features of the signal, obtaining a signal that conforms to physiological characteristics. The baseline drift of the signal that conforms to physiological characteristics is finally corrected to eliminate the remaining low-frequency drift components. The correction uses a method combining high-pass filtering and polynomial fitting to achieve zero-phase filtering and avoid introducing phase distortion, obtaining a stable zero-baseline signal.

[0187] The stable zero-baseline signal is subjected to abnormal waveform detection and repair to identify and correct possible sudden abnormalities (such as spikes, steps, etc.). The detection algorithm is based on the statistical characteristics and morphological feature analysis of abnormal points, and machine learning methods (such as support vector machines or random forests) are used for abnormal waveform classification. For the identified abnormal waveforms, adaptive interpolation or template replacement methods are used for repair to maintain the continuity and consistency of the signal, obtaining a corrected signal. The corrected signal is subjected to bandwidth optimization and smoothing processing to precisely control the frequency band characteristics of the signal and remove the remaining high-frequency noise and irregular fluctuations. The processing uses a zero-phase FIR filter, and the filter parameters are optimized according to the spectral characteristics of the target bioelectric signal to ensure effective noise suppression while maximizing the retention of useful information, obtaining the final smooth signal.

[0188] The morphological matching verification of the final smooth signal with the standard bioelectric template is carried out to evaluate the authenticity and accuracy of the signal. The dynamic time warping (DTW) algorithm is used in the matching process to solve the possible time scale variation problems between different signals and accurately evaluate the waveform similarity. At the same time, the position error and amplitude error of key feature points (such as wave peaks, wave valleys, zero-crossing points, etc.) are calculated to comprehensively evaluate the signal quality. The signal that passes the verification is confirmed as a high-quality pure bioelectric signal and can be used for subsequent physiological parameter extraction and medical diagnostic analysis.

[0189] In this embodiment, the detailed implementation steps of step 8 include:

[0190] Calculating the signal-to-noise ratio, effective bandwidth, and waveform distortion of the pure bioelectric signal to obtain a multidimensional quality evaluation index, and constructing a comprehensive quality function based on the multidimensional quality evaluation index to obtain a signal quality index;

[0191] Establishing a mapping relationship between the signal quality index and key system parameters to obtain a parameter optimization model, and designing a feedback control algorithm based on the parameter optimization model to obtain a system parameter adjustment strategy;

[0192] Based on the system parameter adjustment strategy, the gain parameter G of the second-stage variable gain amplifier is updated in real time to obtain dynamic optimized gain control, and the bandwidth parameter B of the third-stage dynamic bandpass filter is adaptively adjusted according to the signal characteristics to obtain optimized frequency band setting;

[0193] Feeding back the dynamic optimization gain control and the optimized frequency band setting to a hardware circuit to obtain a closed-loop self-optimization structure, and dynamically adjusting the electrode array channel configuration based on the closed-loop self-optimization structure to obtain an optimized acquisition path;

[0194] By integrating the synergistic effect of the closed-loop self-optimization structure and the optimized acquisition path, adaptive optimization of the entire biopotential signal acquisition process is achieved, and a high signal-to-noise ratio biopotential signal output is obtained.

[0195] In this embodiment, a number of quality assessment indicators are first calculated for the pure bioelectric signal. The signal-to-noise ratio (SNR) is calculated using the signal energy to noise energy ratio method, and is estimated by spectral analysis or reference signal comparison. The effective bandwidth is determined by calculating the frequency range where the signal energy is concentrated, and is defined as the minimum bandwidth that contains 90% of the signal energy. The waveform distortion is quantified by calculating the root mean square error (RMSE) or structural similarity index (SSIM) between the signal and the ideal template. These indicators are combined to form a multidimensional quality assessment indicator that comprehensively reflects different aspects of signal quality. A comprehensive quality function is constructed based on the multidimensional quality assessment indicators, and a weighted summation or fuzzy logic method is used to map the multidimensional indicators into a single comprehensive evaluation value to obtain a signal quality index (SQI), which usually ranges from 0 to 1. The higher the value, the better the signal quality.

[0196] Establish the mapping relationship between the signal quality index and the key system parameters, and analyze the influence of different parameter settings on the signal quality. The key parameters include amplifier gain G, filter bandwidth B, center frequency fc, electrode configuration, etc. The mapping relationship is expressed by a mathematical model, which can be a regression model based on experimental data or an analytical model based on theoretical analysis. The model can predict the influence of parameter changes on the signal quality, provide a theoretical basis for parameter optimization, and obtain a parameter optimization model. Design a feedback control algorithm according to the parameter optimization model. This algorithm can calculate the optimal parameter adjustment direction and amplitude according to the current signal quality index. The control algorithm adopts methods such as PID control, model predictive control or reinforcement learning, and has the characteristics of fast convergence and stable control, so as to obtain a system parameter adjustment strategy.

[0197] Based on the system parameter adjustment strategy, the gain parameter G of the second-stage variable gain amplifier is updated in real time. The gain adjustment formula is:

[0198] G(t + 1) = G(t) + ΔG; where G(t) is the gain value at the current time t, G(t + 1) is the gain value at the next time, and ΔG is calculated according to the control algorithm, generally in the range of ±1dB to ±10dB. The gain adjustment is realized by a digital potentiometer or a programmable gain amplifier, and the response time <10ms, ensuring that the system can quickly adapt to the signal amplitude change, and obtain a dynamic optimized gain control. Similarly, the bandwidth parameter B and the center frequency fc of the third-stage dynamic band-pass filter are adaptively adjusted according to the signal characteristics to optimize the frequency response characteristics of the filter. The bandwidth adjustment range is generally between ±5Hz and ±20Hz, and the center frequency adjustment range is between ±2Hz and ±10Hz. The adjustment is realized by digital control capacitor or switched capacitor filter technology, and an optimized frequency band setting is obtained.

[0199] The parameters of the dynamic optimized gain control and the optimized frequency band setting are fed back to the hardware circuit in real time through a digital control interface to realize the dynamic adjustment of the system parameters and form a closed-loop self-optimizing structure. The update frequency of the feedback control loop is 10 - 50Hz, meeting the real-time processing requirements. Based on the closed-loop self-optimizing structure, the dynamic adjustment of the electrode array channel configuration is further realized. The optimal acquisition electrode combination is automatically selected according to the signal quality of each channel, and the signal path is dynamically switched to obtain an optimized acquisition path. The channel optimization algorithm considers the electrode-skin contact impedance, environmental interference distribution and target signal characteristics, comprehensively evaluates the signal quality of each channel, and selects the optimal channel combination.

[0200] By integrating the synergistic effects of the closed-loop self-optimization structure and the optimized acquisition path, adaptive optimization throughout the entire process of bioelectric potential signal acquisition is achieved. The system can adjust the parameters of each link in real time according to environmental changes, human body state changes, and signal characteristic changes to maintain the best working state. The optimization process takes into account the balance of multiple objectives, ensuring system stability and power consumption efficiency while improving the signal-to-noise ratio. Finally, a bioelectric potential signal output with a high signal-to-noise ratio is obtained, with the signal-to-noise ratio increased by 10 - 20 dB, the effective bandwidth precisely matching the target signal characteristics, and the waveform distortion reduced by more than 50%, providing a high-quality data basis for subsequent bioelectric signal analysis and medical diagnosis.

[0201] The low-noise bioelectric potential signal acquisition and processing method in the embodiments of the present application was described above. Next, the low-noise bioelectric potential signal acquisition system in the embodiments of the present application will be described. Please refer to Figure 3 , an embodiment of the low-noise bioelectric potential signal acquisition system in the embodiments of the present application includes:

[0202] An electrode acquisition module, configured to acquire bioelectric potential signals from the human body through an electrode array to obtain raw bioelectric signals;

[0203] A multi-stage adaptive filtering and amplification module, configured to input the raw bioelectric signals into a multi-stage adaptive filtering and amplification circuit to obtain an amplified and filtered analog signal;

[0204] An analog-to-digital conversion module, configured to perform analog-to-digital conversion on the amplified and filtered analog signal to obtain a digitized bioelectric signal;

[0205] A wavelet transform denoising module, configured to perform adaptive wavelet transform denoising processing on the digitized bioelectric signal to obtain a primary denoised signal;

[0206] An empirical mode decomposition module, configured to perform empirical mode decomposition on the primary denoised signal to obtain a multi-level intrinsic mode function set;

[0207] A feature extraction module, configured to extract the intrinsic mode functions in the characteristic frequency band from the multi-level intrinsic mode function set to obtain a characteristic intrinsic mode function set;

[0208] A signal reconstruction module, configured to perform signal reconstruction based on the characteristic intrinsic mode function set to obtain a pure bioelectric signal;

[0209] A quality assessment and parameter optimization module, configured to perform real-time quality assessment on the pure bioelectric signal to obtain a signal quality index; dynamically adjust system parameters based on the signal quality index to optimize the bioelectric potential signal acquisition process and obtain a bioelectric potential signal output with a high signal-to-noise ratio.

[0210] The present invention improves the signal-to-noise ratio and fidelity of signals, enabling weak bioelectrical signals to be accurately captured in complex environments. It effectively overcomes problems such as signal instability, feature loss, and poor adaptability in traditional technologies, and achieves high-quality processing of non-stationary bio-signals. The closed-loop self-optimization mechanism established by the method can adjust system parameters in real time according to signal quality, enabling the system to have excellent environmental adaptability and individual compatibility, and providing a stable and reliable signal basis for different application scenarios. By retaining the key physiological features of the signal while effectively suppressing various interferences, the accuracy and reliability of bioelectrical signal analysis are significantly improved.

[0211] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

[0212] It should be noted that in this article, the terms "including", "comprising" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements but also includes other elements not explicitly listed, or further includes elements inherent to such a process, method, article or device. Without further limitations, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including the said element.

[0213] In the description of the present invention, it should be understood that the terms "first", "second", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0214] In the description of the present invention, unless otherwise specified, the meaning of "a plurality" is two or more.

[0215] In the description of the present invention, the meaning of "several" is one or more, and the meaning of "a large number" is two or more.

[0216] In the description of this specification, the description referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0217] For the formulas in this specification, the dimensional units are removed and only the numerical values are calculated. The formula is obtained by collecting a large amount of data and performing software simulation to get a formula that is closest to the actual situation. The preset parameters and threshold selection in the formula are set by those skilled in the art according to the actual situation.

[0218] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the claims and their equivalents.

Claims

1. A method for collecting and processing low-noise biopotential signals, characterized in that, Including: Step 1: Collect bioelectric potential signals from the human body through an electrode array to obtain original bioelectric signals; Step 2: Input the original bioelectric signals into a multi-stage adaptive filtering and amplification circuit to obtain amplified and filtered analog signals; Step 3: Perform analog-to-digital conversion on the amplified and filtered analog signals to obtain digitalized bioelectric signals; Step 4: Perform adaptive wavelet transform denoising processing on the digitalized bioelectric signals to obtain primary denoised signals; Step 5: Perform empirical mode decomposition on the primary denoised signals to obtain a multi-level intrinsic mode function set; Step 6: Extract the intrinsic mode functions of the characteristic frequency band from the multi-level intrinsic mode function set to obtain a characteristic intrinsic mode function set; Step 7: Perform signal reconstruction based on the characteristic intrinsic mode function set to obtain pure bioelectric signals; Step 8: Perform real-time quality assessment on the pure bioelectric signals to obtain signal quality indicators; Dynamically adjust system parameters based on the signal quality indicators to optimize the bioelectric signal acquisition process and obtain high signal-to-noise ratio bioelectric potential signal output.

2. The low-noise bioelectric potential signal acquisition and processing method according to claim 1, wherein The process of collecting bioelectric potential signals from the human body through an electrode array to obtain original bioelectric signals includes: Perform nano-material modification on the electrode surface to obtain low-impedance bioelectrodes, and arrange the low-impedance bioelectrodes to form an adaptive structure electrode array to obtain a multi-channel acquisition network; Automatically measure the electrode-skin interface impedance of the multi-channel acquisition network to obtain interface contact quality data, and optimize channel selection according to the interface contact quality data to obtain an optimized acquisition channel group; Real-time collect the changes in bioelectric potential on the human body surface through the optimized acquisition channel group to obtain multi-channel original signals, and suppress inter-channel interference on the multi-channel original signals to obtain a preliminary isolated signal; Perform differential amplification preprocessing on the preliminary isolated signal to obtain a differential enhanced signal, and perform preliminary baseline drift correction on the differential enhanced signal to obtain a stable baseline signal; Transmit the stable baseline signal to the subsequent processing circuit through a shielded transmission line to obtain original bioelectric signals.

3. The low-noise bioelectric potential signal acquisition and processing method according to claim 2, wherein The process of inputting the original bioelectric signals into a multi-stage adaptive filtering and amplification circuit to obtain amplified and filtered analog signals includes: Input the original bioelectric signals into a first-stage low-pass filter to obtain a first-stage filtered signal with high-frequency interference removed. The first-stage low-pass filter adopts a second-order Butterworth structure and its cut-off frequency is adjustable; Input the first-stage filtered signal into a second-stage variable gain amplifier to obtain an amplified signal with optimized dynamic range. The gain parameter G of the second-stage variable gain amplifier is dynamically adjusted through real-time feedback control; Input the amplified signal with optimized dynamic range into a third-stage dynamic band-pass filter to obtain a second-stage filtered signal with optimized frequency band. The center frequency fc and bandwidth parameter B of the third-stage dynamic band-pass filter are adaptively adjusted according to the signal characteristics; Perform adaptive notch filtering for power frequency noise on the second-stage filtered signal with optimized frequency band to obtain a filtered signal with power frequency interference suppressed. The adaptive notch filtering for power frequency noise uses a phase-locked loop tracking technology to achieve compensation for power frequency drift; Output buffer and impedance matching are performed on the filtered signal that suppresses power frequency interference to obtain an amplified and filtered analog signal.

4. The low-noise biological potential signal acquisition and processing method according to claim 3, characterized in that The analog-to-digital conversion of the amplified and filtered analog signal to obtain a digitized bioelectrical signal includes: Signal conditioning is performed on the amplified and filtered analog signal to obtain a standardized signal that meets the conversion range, and sample and hold is performed on the standardized signal to obtain a discrete-time sampling point sequence; A high-precision ADC is used to perform quantization encoding on the discrete-time sampling point sequence to obtain an initial digital code stream. The sampling rate of the high-precision ADC is dynamically adjusted according to the signal spectrum characteristics, and the resolution is not less than 16 bits; Quantization noise shaping is performed on the initial digital code stream to obtain an optimized digital signal, and digital filtering anti-aliasing processing is performed on the optimized digital signal to obtain an anti-aliasing digital signal; Timing calibration is performed on the anti-aliasing digital signal to obtain a synchronous digital signal, and data format conversion is performed on the synchronous digital signal to obtain digital data in a standard format; The digital data in the standard format is subjected to data stream control through a buffer management mechanism to obtain a digitized bioelectrical signal.

5. The low-noise bioelectric potential signal acquisition and processing method according to claim 4, wherein The adaptive wavelet transform denoising process is performed on the digitized bioelectrical signal to obtain a primary denoised signal, including: Signal characteristic analysis is performed on the digitized bioelectrical signal to obtain a signal spectrum feature map, and an optimal wavelet basis function is selected according to the signal spectrum feature map to obtain an adapted wavelet basis; The digitized bioelectrical signal is subjected to multi-scale wavelet decomposition using the adapted wavelet basis to obtain a multi-layer wavelet coefficient set, and threshold analysis is performed on the multi-layer wavelet coefficient set to obtain a noise threshold distribution; An adaptive threshold function is designed according to the noise threshold distribution to obtain an optimized threshold strategy, and the optimized threshold strategy is applied to perform soft threshold processing on the wavelet coefficients to obtain denoised wavelet coefficients; A signal singularity detection algorithm is applied to the denoised wavelet coefficients to obtain a singularity protection mapping, and coefficient correction is performed according to the singularity protection mapping to obtain a wavelet coefficient set that preserves features; Inverse wavelet transform reconstruction is performed using the wavelet coefficient set that preserves features to obtain a primary denoised signal.

6. The low-noise biological potential signal acquisition and processing method according to claim 5, wherein The empirical mode decomposition is performed on the primary denoised signal to obtain a multi-level intrinsic mode function set, including: Extreme point detection is performed on the primary denoised signal to obtain a signal extreme value distribution map, and upper and lower envelope lines are constructed based on the signal extreme value distribution map to obtain an envelope pair; The mean envelope line is calculated according to the envelope pair to obtain a local mean function, and the primary denoised signal is subtracted from the local mean function to obtain a candidate intrinsic mode function; Intrinsic property testing is performed on the candidate intrinsic mode function to obtain an intrinsic property evaluation result. If the intrinsic property evaluation result does not meet the IMF condition, the envelope calculation process is repeated until the IMF condition is met to obtain the first-layer intrinsic mode function; The primary denoised signal is subtracted from the first-layer intrinsic mode function to obtain a first residual signal, and the empirical mode decomposition process is repeatedly performed on the first residual signal to obtain subsequent intrinsic mode functions; Combine all the decomposed intrinsic mode functions and the final residual signal to obtain a complete multi-level intrinsic mode function set.

7. The low-noise biological potential signal acquisition and processing method according to claim 6, wherein Extracting the intrinsic mode functions of the characteristic frequency band from the multi-level intrinsic mode function set to obtain a characteristic intrinsic mode function set, including: Perform Hilbert transform on each IMF in the multi-level intrinsic mode function set to obtain the instantaneous frequency distribution, and calculate the spectral center and bandwidth of each IMF according to the instantaneous frequency distribution to obtain the IMF spectral characteristic table; Based on the IMF spectral characteristic table, analyze the characteristic frequency band of the bioelectrical potential signal to obtain the target frequency band range, and screen the relevant IMFs according to the target frequency band range to obtain the preliminary selected IMF set; Perform cross-correlation analysis on the preliminary selected IMF set to obtain the IMF correlation matrix, and remove redundant components according to the IMF correlation matrix to obtain the simplified IMF set; Calculate the energy contribution degree of each IMF in the simplified IMF set to obtain the information quantity evaluation result, and perform importance ranking according to the information quantity evaluation result to obtain the ranked IMF set; Based on the ranked IMF set, perform adaptive weight allocation to obtain a weighted IMF combination scheme, and construct a characteristic intrinsic mode function set according to the weighted IMF combination scheme.

8. The low-noise biological potential signal acquisition and processing method according to claim 7, characterized in that Based on the characteristic intrinsic mode function set, perform signal reconstruction to obtain a pure bioelectrical signal, including: Perform non-linear phase correction on the characteristic intrinsic mode function set to obtain a phase-aligned IMF group, and design an optimal reconstruction weight vector for the phase-aligned IMF group to obtain an adaptive fusion strategy; Based on the adaptive fusion strategy, perform weighted superposition on the IMFs to obtain a preliminary reconstructed signal, and perform morphological feature enhancement on the preliminary reconstructed signal to obtain a bioelectrical signal with prominent features; Perform physiological model constraint processing on the bioelectrical signal with prominent features to obtain a signal that conforms to physiological characteristics, and perform final correction of baseline drift on the signal that conforms to physiological characteristics to obtain a stable zero-baseline signal; Perform abnormal waveform detection and repair on the stable zero-baseline signal to obtain a corrected signal, and perform bandwidth optimization and smoothing processing on the corrected signal to obtain a final smooth signal; Perform morphological matching verification on the final smooth signal and the standard bioelectrical template to obtain a pure bioelectrical signal.

9. The low-noise biological potential signal acquisition and processing method according to claim 8, wherein Performing real-time quality assessment on the pure bioelectrical signal to obtain a signal quality index, and dynamically adjusting system parameters based on the signal quality index to optimize the bioelectrical signal acquisition process to obtain a high signal-to-noise ratio bioelectrical potential signal output, including: Calculate the signal-to-noise ratio, effective bandwidth, and waveform distortion degree of the pure bioelectrical signal to obtain a multi-dimensional quality assessment index, and construct a comprehensive quality function according to the multi-dimensional quality assessment index to obtain a signal quality index; Establish a mapping relationship between the signal quality index and the key system parameters to obtain a parameter optimization model, and design a feedback control algorithm according to the parameter optimization model to obtain a system parameter adjustment strategy; Based on the system parameter adjustment strategy, the gain parameter G of the second-stage variable gain amplifier is updated in real time to obtain dynamic optimized gain control, and the bandwidth parameter B of the third-stage dynamic band-pass filter is adaptively adjusted according to the signal characteristics to obtain optimized frequency band settings; The dynamic optimized gain control and the optimized frequency band settings are fed back to the hardware circuit to obtain a closed-loop self-optimized structure, and based on the closed-loop self-optimized structure, the dynamic adjustment of the electrode array channel configuration is realized to obtain an optimized acquisition path; By integrating the synergistic effects of the closed-loop self-optimized structure and the optimized acquisition path, the adaptive optimization of the whole process of biopotential signal acquisition is realized to obtain a high signal-to-noise ratio biopotential signal output.

10. A low-noise bioelectric potential signal acquisition system for implementing the low-noise bioelectric potential signal acquisition and processing method according to any one of claims 1 to 9, characterized in that, It includes: An electrode acquisition module for acquiring biopotential signals from the human body through an electrode array to obtain original bioelectric signals; A multi-stage adaptive filtering and amplification module for inputting the original bioelectric signals into a multi-stage adaptive filtering and amplification circuit to obtain amplified and filtered analog signals; An analog-to-digital conversion module for performing analog-to-digital conversion on the amplified and filtered analog signals to obtain digitalized bioelectric signals; A wavelet transform denoising module for performing adaptive wavelet transform denoising processing on the digitalized bioelectric signals to obtain primary denoised signals; An empirical mode decomposition module for performing empirical mode decomposition on the primary denoised signals to obtain a multi-level intrinsic mode function set; A feature extraction module for extracting the intrinsic mode functions of the characteristic frequency bands from the multi-level intrinsic mode function set to obtain a characteristic intrinsic mode function set; A signal reconstruction module for performing signal reconstruction based on the characteristic intrinsic mode function set to obtain pure bioelectric signals; A quality evaluation and parameter optimization module for performing real-time quality evaluation on the pure bioelectric signals to obtain signal quality indicators; dynamically adjusting system parameters based on the signal quality indicators to optimize the biopotential signal acquisition process to obtain a high signal-to-noise ratio biopotential signal output.

Images