A sparrow search-based adaptive noise reduction method, device, medium and product
By optimizing the weight coefficients of the adaptive filter using the sparrow search algorithm, the problem of severe noise interference in surface electromyography signals by traditional filters is solved, and a more efficient noise suppression effect is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI UNIV
- Filing Date
- 2024-03-06
- Publication Date
- 2026-07-31
AI Technical Summary
Traditional adaptive filters struggle to achieve high filtering performance when dealing with complex mixed noise problems, especially in surface electromyography signals, where noise interference severely affects signal quality.
The sparrow search algorithm is used to iteratively update the weight coefficients of the adaptive filter. By constructing an error function and a fitness function, the noise reduction performance of the filter is optimized.
It improves the denoising capability of the adaptive filter, enhances signal quality, reduces noise interference, and improves the signal-to-noise ratio and mean square error.
Smart Images
Figure CN118227964B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of electromyography signal processing technology, and in particular to an adaptive noise reduction method, device, medium, and product based on sparrow search. Background Technology
[0002] Surface electromyography (EMG) signals are bioelectrical signals formed by the superposition of action potentials generated by numerous muscle fibers on the surface of human skin during muscle activity. They have significant practical value in clinical medicine and automated robotics. The signal quality of the surface EMG signals acquired by EMG electrodes is a major factor in ensuring the effectiveness of practical applications. When directly acquiring surface EMG signals, the high skin impedance makes the signal transmission highly susceptible to power frequency noise, resulting in poor signal quality. Furthermore, changes in the relative position of the EMG electrodes to the skin surface can introduce low-frequency noise such as motion artifacts. Additionally, if the acquisition location is close to the heart, the acquired signal will be affected by electrocardiogram (ECG) signals. Therefore, software filtering of the acquired surface EMG signals is crucial.
[0003] An adaptive filter is a recursive filter that correlates two related signals in a feedback manner, with its weights varying according to changes in ambient noise. Therefore, adaptive filters can be used to eliminate dynamically changing random noise in the frequency domain of electromyography (EMG) signals, as well as electric field noise in the same frequency band as EMG signals. Adaptive filters can be implemented using FIR or IIR methods, minimizing the cost function through the steepest descent method, where the filter weights or coefficients act as features in a linear regression problem. However, traditional adaptive filters use a fixed step size to calculate the filter weight coefficient matrix, making it difficult to achieve high filtering performance when dealing with complex mixed noise problems. Summary of the Invention
[0004] The purpose of this invention is to provide an adaptive noise reduction method, device, medium, and product based on sparrow search, which can improve the noise reduction capability and performance of filters.
[0005] To achieve the above objectives, the present invention provides the following solution:
[0006] An adaptive noise reduction method based on sparrow search, the method comprising:
[0007] Acquire electromyographic (EMG) signal data and adaptive filter parameter data; the EMG signal data includes: raw acquired signal data and environmental noise reference signal data; the adaptive filter parameter data includes: order, weight coefficients, and number of iterations.
[0008] The error function corresponding to the error signal is determined based on the electromyographic signal data and the adaptive filter parameter data.
[0009] The fitness function is determined based on the error function.
[0010] The Sparrow Search algorithm is used to iteratively update the weight coefficients to meet the iteration stopping condition, resulting in updated weight coefficients. The iteration stopping condition is that the current iteration number reaches a set iteration number or the fitness function value is minimized at the current iteration number. The updated weight coefficients are used to improve the filtering and noise reduction performance of the adaptive filter.
[0011] Optionally, the expression for the error function specifically includes:
[0012] e(n)=d(n)-y(n)=d(n)-W T (n)r(n);
[0013] Where e(n) is the error function; d(n) is the array storing the original acquired signal data; y(n) is the output value of the adaptive filter; W(n) is the weighting coefficient; r(n) is the array storing the environmental noise reference signal data; T is the vector transpose; and n is the number of iterations.
[0014] Optionally, the expression for the fitness function specifically includes:
[0015]
[0016] Where f(n) is the fitness function; G i (n) represents the mean square error; n represents the number of iterations; d(n) is an array storing the original acquired signal data; W(n) represents the weighting coefficient; T represents the vector transpose; M represents the total number of input signal samples based on electromyographic signal data; i represents the sequence number of the input signal sample; r(n) is an array storing the environmental noise reference signal data.
[0017] Optionally, a sparrow search method is used to iteratively update the weight coefficients to meet the iteration stopping condition, resulting in updated weight coefficients, specifically including:
[0018] Initial information data for determining a sparrow population; the initial information data includes: population size and spatial dimensions; the sparrow population is defined as electromyographic signal data;
[0019] The sparrow population is sorted by position according to the fitness function to obtain sparrow position information; the sparrow position information is a weighting coefficient.
[0020] Using a sparrow search method, the sparrow population was divided into joiners, discoverers, and vigilant sparrows;
[0021] With the goal of satisfying the iteration stopping condition, the positions of the joiner, discoverer, and vigilant are iteratively updated to obtain the updated sparrow position information; the updated sparrow positions are used to perform global optimization of the adaptive filter for noise reduction processing.
[0022] Optionally, the iterative update process of the discoverer's location specifically includes:
[0023]
[0024] in, This represents the position of the discoverer in the (t+1)th iteration. Let be the position of the discoverer in the t-th iteration; T1 is the set iteration number of the sparrow search algorithm; α is a uniformly random number between (0,1]; Q is a random number following a standard normal distribution; L1 is a 1×d matrix of all 1s; R2 is the warning value; S T is a safe value; i is the sequence number of the input signal sample.
[0025] Optionally, the iterative update process of the joiner's position includes:
[0026]
[0027] in, This represents the position of the joiner in the (t+1)th iteration. Let be the sparrow's worst position in the d-th dimension during the t-th iteration of the population; This represents the position of the joiner at the t-th iteration. Let A be the optimal position of the sparrow in the d-th dimension at the (t+1)-th iteration of the population; A is a 1×d matrix; N is the population size; A + A is an intermediate quantity; + =A T (A·A T ) -1 .
[0028] Optionally, the position update process of the alerter specifically includes:
[0029]
[0030] in, The position of the vigilant at the (t+1)th iteration; Let be the optimal position of the sparrow in the d-th dimension during the t-th iteration of the population; β is the step size control parameter that follows a normal distribution of random numbers; f represents the position of the vigilant at the t-th iteration. i f is the fitness value of the i-th sparrow; gThe current global optimal fitness value; K is the step size control parameter between [-1, 1]; f w is the current global worst fitness value; e is a constant.
[0031] A computer device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the computer program to implement the steps of the above-described adaptive noise reduction method based on sparrow search.
[0032] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described adaptive noise reduction method based on sparrow search.
[0033] A computer program product includes a computer program that, when executed by a processor, implements the steps of the above-described adaptive noise reduction method based on sparrow search.
[0034] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:
[0035] This invention discloses an adaptive noise reduction method, device, medium, and product based on sparrow search. It acquires electromyography (EMG) signal data and adaptive filter parameter data; determines the error function corresponding to the error signal based on the EMG signal data and adaptive filter parameter data; determines the fitness function based on the error function; employs a sparrow search algorithm, with the goal of satisfying the iteration stopping condition, to iteratively update the weight coefficients, obtaining the updated weight coefficients; the iteration stopping condition is that the current iteration number reaches a set iteration number or the fitness function value at the current iteration number is minimized; the updated weight coefficients are used to improve the filtering and noise reduction performance of the adaptive filter; thereby, the filter's noise reduction capability and performance are improved. Attached Figure Description
[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0037] Figure 1 A schematic diagram of the adaptive noise reduction method based on sparrow search provided in an embodiment of the present invention;
[0038] Figure 2 This is an architecture diagram of the adaptive active noise reduction method provided in an embodiment of the present invention;
[0039] Figure 3This is a flowchart illustrating the operation of the adaptive active noise reduction method provided in this embodiment of the invention.
[0040] Figure 4 A schematic diagram of the original acquisition signal containing mixed noise, the reference signal containing mixed noise, and the surface electromyography signal after noise reduction, provided for embodiments of the present invention;
[0041] Figure 5 A schematic diagram showing the signal-to-noise ratio comparison of surface electromyography signals after filtering raw signals with different input signal-to-noise ratios using different methods, provided in an embodiment of the present invention;
[0042] Figure 6 A schematic diagram showing the mean square error comparison of surface electromyography signals after filtering raw signals with different input signal-to-noise ratios using different methods, as provided in an embodiment of the present invention;
[0043] Figure 7 This diagram illustrates the convergence speed using different methods, as provided in embodiments of the present invention. Detailed Implementation
[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0045] To address the issue of poor signal quality after filtering due to the difficulty in achieving high convergence accuracy in mixed noise, as proposed in foreign literature (Reference 1, translated as "A Novel Adaptive Filtering Method for Electromyography Signals Using the Gray Wolf Optimization Algorithm") and (Reference 2, translated as "Eliminating Noise in Surface Electromyography via Enhanced Adaptive Filtering Technique Based on Evolutionary Computation"), this invention proposes an adaptive active denoising method based on sparrow search. This method improves the denoising capability of the filter by mitigating the premature convergence problem of traditional evolutionary algorithms.
[0046] The purpose of this invention is to provide an adaptive noise reduction method, device, medium, and product based on sparrow search, which aims to improve the noise reduction capability and performance of filters.
[0047] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0048] Example 1
[0049] like Figure 1 As shown in this embodiment, an adaptive noise reduction method based on sparrow search is provided. The method includes:
[0050] Step 100: Acquire electromyographic signal data and adaptive filter parameter data. The electromyographic signal data includes: raw acquired signal data and environmental noise reference signal data; the adaptive filter parameter data includes: order, weighting coefficients, and number of iterations.
[0051] Step 200: Determine the error function corresponding to the error signal based on the electromyography signal data and the adaptive filter parameter data.
[0052] Step 300: Determine the fitness function based on the error function.
[0053] Step 400: Employ the Sparrow Search algorithm to iteratively update the weight coefficients, aiming to satisfy the iteration stopping condition, to obtain the updated weight coefficients. The iteration stopping condition is that the current iteration count reaches a set number of iterations or the fitness function value is minimized at the current iteration count. The updated weight coefficients are used to improve the filtering and noise reduction performance of the adaptive filter.
[0054] The expression for the error function specifically includes:
[0055] e(n)=d(n)-y(n)=d(n)-W T (n)r(n).
[0056] Where e(n) is the error function; d(n) is the array storing the original acquired signal data; y(n) is the output value of the adaptive filter; W(n) is the weighting coefficient; r(n) is the array storing the environmental noise reference signal data; T is the vector transpose; and n is the number of iterations.
[0057] The expression for the fitness function specifically includes:
[0058]
[0059] Where f(n) is the fitness function; G i (n) represents the mean square error; n represents the number of iterations; d(n) is an array storing the original acquired signal data; W(n) represents the weighting coefficient; T represents the vector transpose; M represents the total number of input signal samples based on electromyographic signal data; i represents the sequence number of the input signal sample; r(n) is an array storing the environmental noise reference signal data.
[0060] In one embodiment, a sparrow search method is used to iteratively update the weight coefficients to meet the iteration stopping condition, resulting in updated weight coefficients. Specifically, this includes:
[0061] Initial information data for determining the sparrow population was obtained. This initial information data included population size and spatial dimensions. The sparrow population data was obtained from electromyographic signal data.
[0062] The sparrow population is sorted by position according to the fitness function to obtain sparrow position information; the sparrow position information is the weight coefficient.
[0063] Using a sparrow search method, sparrow populations were divided into joiners, discoverers, and vigilant individuals.
[0064] With the goal of satisfying the iteration stopping condition, the positions of the joiner, discoverer, and vigilant are iteratively updated to obtain the updated sparrow position information; the updated sparrow positions are used to perform global optimization of the adaptive filter for noise reduction.
[0065] The iterative update process of the discoverer's location specifically includes:
[0066]
[0067] in, This represents the position of the discoverer in the (t+1)th iteration. Let be the position of the discoverer in the t-th iteration; T1 is the set iteration number of the sparrow search algorithm; α is a uniformly random number between (0,1]; Q is a random number following a standard normal distribution; L1 is a 1×d matrix of all 1s; R2 is the warning value; S T is a safe value; i is the sequence number of the input signal sample.
[0068] The iterative update process of the joiner's position includes:
[0069]
[0070] in, This represents the position of the joiner in the (t+1)th iteration. Let be the sparrow's worst position in the d-th dimension during the t-th iteration of the population; This represents the position of the joiner at the t-th iteration. Let A be the optimal position of the sparrow in the d-th dimension at the (t+1)-th iteration of the population; A is a 1×d matrix; N is the population size; A + A is an intermediate quantity; + =A T (A·A T ) -1 .
[0071] The location update process of the alerter includes the following:
[0072]
[0073] in, The position of the vigilant at the (t+1)th iteration; Let be the optimal position of the sparrow in the d-th dimension during the t-th iteration of the population; β is the step size control parameter that follows a normal distribution of random numbers; f represents the position of the vigilant at the t-th iteration. i f is the fitness value of the i-th sparrow; g The current global optimal fitness value; K is the step size control parameter between [-1, 1]; f w is the current global worst fitness value; e is a constant.
[0074] like Figure 2 and Figure 3 As shown, in practical applications, the specific operation steps of the method provided in this embodiment of the invention are as follows:
[0075] S1. Initialize parameters and load signals.
[0076] S11. Initialize the adaptive filter order L and the filter weight coefficients W(n) = [w0(n), w1(n), ..., w L (n)] T L represents the filter order, T represents the vector transpose, and n represents the number of iterations.
[0077] S12. Initialize the sparrow's position x i,j .
[0078] x i,j =x l +u(0,1)×(x u -x l ).
[0079] Where, x l With x u Let x represent the upper and lower boundaries of the sparrow's position, respectively. i,d Let represent the position of the i-th sparrow in the d-th dimension, where i = 1, 2, ..., N, and N represents the total number of sparrows in the population.
[0080] S13. Load the original acquired signal d(k) containing mixed noise and the environmental noise reference signal r(k) containing mixed noise.
[0081] S2. Mixed noise cancellation.
[0082] S21. Store the original acquired signal d(k) containing mixed noise into array d(n).
[0083] d(n)=[d(kL+1),d(kL+2),...,d(kL+L)] T .
[0084] Where L represents the filter order, T represents the vector transpose, k represents the index of the data in the time domain, and n represents the number of iterations.
[0085] S22. Store the ambient noise reference signal r(k) containing mixed noise into array r(n).
[0086] r(n)=[r(nL+1),r(nL+2),...,r(nL+L)] T .
[0087] S23. Calculate the output value y(n) of the linear filter.
[0088]
[0089] Where W(n)=[w0(n),w1(n),...,w L (n)] T These are the weighting coefficients of the filter.
[0090] S24. Calculate the error signal e(n), which is not only the filtered surface electromyography signal, but also the cost function used to calculate and evaluate the mean square error.
[0091] e(n)=d(n)-y(n)=d(n)-W T (n)r(n).
[0092] S3. Construction of the cost function for the sparrow search and updating of the filter weight coefficients.
[0093] S31. Calculate the sparrow fitness f x (n).
[0094] S31 includes the following steps:
[0095] S311. Create the fitness matrix f of the sparrow population. x .
[0096]
[0097] Where N represents the sparrow population size in d-dimensional space, f x This represents the fitness of the sparrow population in dimension x.
[0098] S312. The random solution set generated by the sparrow search contains all the weight coefficients of the adaptive filter. Calculate the mean square error G. i (n).
[0099]
[0100] Among them, e ij(n) represents the j-th error of the i-th sample in the n-th iteration, and M represents the total number of input signal samples based on electromyographic signal data.
[0101] S313. Calculate the sparrow fitness f(n).
[0102]
[0103] S32. Sort the fitness and update the sparrow population position. The sparrow with the lowest fitness is in the optimal position Xb, and the sparrow with the highest fitness is in the worst position XW.
[0104] S33. Update the number and position of joiners, discoverers, and vigilants in the sparrow population. The position of the sparrow with fitness is the optimal weight vector W(n) of the filter.
[0105] S33 includes the following steps:
[0106] S331. Update the location of the discoverer.
[0107]
[0108] in, This represents the position of the discoverer in the (t+1)th iteration. Let be the position of the discoverer in the t-th iteration; T1 is the set iteration number of the sparrow search algorithm; α is a uniformly random number between (0,1]; Q is a random number following a standard normal distribution; L1 is a 1×d matrix of all 1s; R2 is the warning value; S T is a safe value; i is the sequence number of the input signal sample.
[0109] R2 and S T (R2∈[0,1],S T ∈[0.5,1]) represent the warning value and the safety value, respectively.
[0110] S332. Update the position of the newcomer.
[0111]
[0112] Where A represents a matrix of dimension 1×d, and each element in the matrix is randomly assigned a value of -1 or 1; Let be the sparrow's worst position in the d-th dimension during the t-th iteration of the population; Let be the optimal position of the sparrow in the d-th dimension during the (t+1)-th iteration of the population.
[0113] S333. Update the location of the vigilant.
[0114]
[0115] Where β is the step size control parameter, a random number following a normal distribution with a mean of 0 and a variance of 1; K, between [-1, 1], is also a step size control parameter, used to represent the direction of the sparrow's movement; e is a very small constant, f i f is the fitness value of the i-th sparrow; g and f w These represent the current global optimal and worst fitness values, respectively.
[0116] S34. Save the current sparrow position and update the global optimal sparrow position, i.e., filter weight W(n) = Xb = [x b,1 ,x b,2 ,...,x b,d ], where x b,d A sparrow representing the best position in each dimension.
[0117] S4. Iterate and update, repeating steps S2 to S3 until the iteration conditions are met to complete the noise reduction.
[0118] Traditional LMS adaptive filters use a fixed step size to update filter weights during training, which makes it difficult to balance convergence speed and accuracy, thus hindering the achievement of good noise reduction results. This invention utilizes a sparrow search to construct a function of the minimum mean square error between the filtered signal and the reference signal, along with the weights, as the cost function for the sparrow search. This reduces the mean square error of the adaptive filter, thereby iteratively updating the weights to improve filtering performance. Compared to traditional adaptive filtering techniques used for surface electromyography (EMG) noise reduction, this invention improves the filter's noise reduction capability by overcoming the premature convergence of traditional evolutionary algorithms.
[0119] To verify the effectiveness of the present invention, a data acquisition experiment was conducted in a specific embodiment, and the results were compared with various adaptive filtering algorithms mentioned in References 1 and 2 in the prior art: Figure 4 (a) in the image represents the surface electromyography signal with a signal-to-noise ratio of -5 dB contaminated by noise. Figure 4 In the diagram, (b) represents the reference noise acquired by the reference channel. The surface electromyography signals filtered using LMS, NLMS, PSO, GWO, and CSO respectively, along with the noise reduction algorithm of this invention, are shown below. Figure 4 As shown in (c)-(h).
[0120] The data acquisition device used was the Quattrocento system developed by OTBioelectronica, and the sampling frequency was set to 2000Hz during the acquisition experiment. The parameter values for the methods in Reference 1, Reference 2, and the method of this invention are shown in Table 1 below.
[0121] Table 1 Parameter Value Table
[0122] Reference 2 <![CDATA[N=25,Samples=2000,x l =-0.1,x u =0.1]]> This invention <![CDATA[N=20,Samples=2000,x l =-0.1,x u =0.1]]>
[0123] In the evaluation criteria for noise reduction effectiveness, the signal-to-noise ratio (SNR) and mean square error (MSE) of the filtered signal are typically used as quantitative evaluation metrics. A higher SNR and lower MSE indicate better performance of the adaptive algorithm and a more effective noise reduction. Figure 5 and Figure 6 The comparisons are as follows: the signal-to-noise ratio and mean square error of surface electromyography signals after filtering the original signals with different input signal-to-noise ratios using the methods in Reference 1, Reference 2, and the method of this invention.
[0124] The quantitative comparison of the denoising results of the methods in References 1 and 2 and the adaptive active denoising method based on sparrow search of the present invention is shown in Table 2 below.
[0125] Table 2 Comparison of Noise Reduction Results
[0126]
[0127]
[0128] As can be seen from the table above, under different input signal-to-noise ratio settings, the signal-to-noise ratio of the adaptive active noise reduction method based on sparrow search of the present invention is higher than that of the methods in Reference 1 and Reference 2, and the root mean square error is lower than that of the methods in Reference 1 and Reference 2. This indicates that the noise reduction effect of the adaptive active noise reduction method based on sparrow search of the present invention is better. Figure 7 The convergence speeds of the methods in Reference 1, Reference 2, and the method of this invention are compared; the adaptive active denoising method based on sparrow search of this invention converges faster. Therefore, the adaptive active denoising method based on sparrow search of this invention has higher computational efficiency and can effectively eliminate mixed noise in surface electromyography signals.
[0129] Example 2
[0130] This invention provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the steps of the adaptive noise reduction method based on sparrow search in Embodiment 1.
[0131] Example 3
[0132] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the adaptive noise reduction method based on sparrow search in Embodiment 1.
[0133] Example 4
[0134] A computer program product includes a computer program that, when executed by a processor, implements the steps of the adaptive noise reduction method based on sparrow search in Embodiment 1.
[0135] Example 5
[0136] A computer device, which may be a database, includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The database stores pending transactions. The I / O interfaces facilitate information exchange between the processor and external devices. The communication interface enables communication with external terminals via a network connection. When executed by the processor, the computer program implements the sparrow search-based adaptive noise reduction method described in Embodiment 1.
[0137] It should be noted that the object information (including but not limited to object device information, object personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this invention are all information and data authorized by the object or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.
[0138] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided by this invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided by this invention may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided by this invention may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.
[0139] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0140] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.
Claims
1. A sparrow search based adaptive noise reduction method, characterized in that, The method includes: Acquire electromyographic (EMG) signal data and adaptive filter parameter data; the EMG signal data includes: raw acquired signal data and environmental noise reference signal data; the adaptive filter parameter data includes: order, weight coefficients, and number of iterations. The error function corresponding to the error signal is determined based on the electromyographic signal data and the adaptive filter parameter data. The fitness function is determined based on the error function. The Sparrow Search algorithm is used to iteratively update the weight coefficients to meet the iteration stopping condition, resulting in updated weight coefficients. The iteration stopping condition is that the current iteration number reaches a set iteration number or the fitness function value is minimized at the current iteration number. The updated weight coefficients are used to improve the filtering and noise reduction performance of the adaptive filter. The expression for the fitness function specifically includes: ； wherein, is a fitness function; is a mean square error; is a number of iterations; is an array for storing original acquisition signal data; is a weight coefficient; is a vector transpose; is a total number of input signal samples based on electromyographic signal data; is a serial number of input signal samples; is an array for storing environmental noise reference signal data; The sparrow search method is used to iteratively update the weight coefficients to meet the iteration stopping condition, resulting in updated weight coefficients. Specifically, this includes: Initial information data for determining a sparrow population; the initial information data includes: population size and spatial dimensions; the sparrow population is represented by electromyographic signal data; The sparrow population is sorted by position according to the fitness function to obtain sparrow position information; the sparrow position information is a weighting coefficient. Using a sparrow search method, the sparrow population was divided into joiners, discoverers, and vigilant sparrows; With the goal of satisfying the iteration stopping condition, the positions of the joiner, discoverer, and vigilant are iteratively updated to obtain the updated sparrow position information; the updated sparrow positions are used to perform global optimization of the adaptive filter for noise reduction processing.
2. The adaptive noise reduction method based on sparrow search according to claim 1, characterized in that, The expression for the error function specifically includes: ； in, It is the error function; An array for storing the raw acquired signal data; This is the output value of the adaptive filter; These are the weighting coefficients; An array for storing environmental noise reference signal data; Transpose of a vector; This represents the number of iterations.
3. The adaptive noise reduction method based on sparrow search according to claim 1, characterized in that, The iterative update process of the discoverer's location specifically includes: ； in, For the first The location of the discoverer in the next iteration; For the first The location of the discoverer in the next iteration; Set the number of iterations for the sparrow search algorithm; A uniformly random number between (0,1); These are random numbers that follow a standard normal distribution. For dimension A matrix of all ones; This is a warning value; This is a safe value; The sequence number of the input signal sample.
4. The adaptive noise reduction method based on sparrow search according to claim 3, characterized in that, The iterative update process of the joiner's position includes: ； in, For the first The position of the joiner in the next iteration; For the population number In the next iteration, the sparrow was at the... The global worst position of the dimension; For the first The position of the joiner in the next iteration; For the population number The optimal position of the sparrow in the d-th dimension at the next iteration; It is a matrix of dimension 1×d; Population size; This is an intermediate quantity; .
5. The adaptive noise reduction method based on sparrow search according to claim 4, characterized in that, The location update process of the alerter includes the following: ； in, For the first The location of the vigilant at the next iteration number; For the population number In the next iteration, the sparrow was at the... The optimal position of the dimension; The step size control parameter for random numbers that follow a normal distribution; For the first The location of the vigilant at the next iteration number; For the first The fitness value of a single sparrow; This is the current globally optimal fitness value; The step size control parameter is between [-1, 1]. This is the current worst-case fitness value globally. It is a constant.
6. A computer device, comprising: The memory and processor contain a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the adaptive noise reduction method based on sparrow search as described in any one of claims 1-5.
7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the adaptive noise reduction method based on sparrow search as described in any one of claims 1-5.
8. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the adaptive noise reduction method based on sparrow search as described in any one of claims 1-5.