Fuse denoising method and system based on adaptive filtering

Abstract

The invention provides a fuze denoising method and system based on adaptive filtering, and relates to the technical field of signal processing. The method comprises the following steps: constructing a filter coefficient iteration model through constant parameter variation according to a variable step size algorithm model; calculating an error value between the input signal data and the output signal data of the filter; under the condition that the error value is not smaller than a preset error limit, iteratively updating a filter coefficient of the filter through a filter coefficient iteration model according to the current error value and a historical error value; and carrying out denoising processing on the signal data by using the updated filter coefficient. According to the method, the convergence speed of the model can be effectively increased, and meanwhile, the steady-state error performance can be relatively good. Moreover, historical error values at different moments before the current secondary error value can be introduced, so that the method has the advantage of high practicability, and can be applied to various different application scenes.

Description

Technical Field

[0001] The present specification relates to the field of signal processing technology, and in particular, to a method and system for fuze denoising based on adaptive filtering. Background Art

[0002] Radio fuzes are easily affected by interference signals. For the denoising of fuze echo signals, there are mainly methods such as wavelet analysis and adaptive filtering. Due to the advantages of LMS adaptive filtering algorithm such as small calculation amount, easy implementation and good stability, it is widely used in system identification, noise cancellation and beamforming. For the traditional variable step size algorithm model, although the characteristics of the variable step size LMS algorithm have been improved to a certain extent, it has the defects of large steady-state error and slow convergence speed. Summary of the invention

[0003] The purpose of this specification is to provide a fuze denoising method based on adaptive filtering, which can overcome the above-mentioned defects of the traditional variable step-size algorithm model.

[0004] The embodiments of this specification are implemented as follows:

[0005] On the one hand, this specification provides a fuze denoising method based on adaptive filtering, which mainly includes:

[0006] According to the variable step size algorithm model, the filter coefficient iteration model is constructed by variableizing the constant parameters;

[0007] Calculate the error value between the filter input signal data and the output signal data;

[0008] Under the condition that the error value is not less than a preset error limit, according to the current error value and the historical error value, the filter coefficient of the filter is iteratively updated through the filter coefficient iteration model;

[0009] The signal data is subjected to denoising processing using the updated filter coefficients.

[0010] On the other hand, the present specification provides a fuze denoising system based on adaptive filtering, which mainly includes:

[0011] A construction module is used to construct a filter coefficient iteration model by variableizing constant parameters according to a variable step size algorithm model;

[0012] A calculation module, used for calculating the error value between the filter input signal data and the output signal data;

[0013] An iteration module, configured to iteratively update the filter coefficients of the filter through the filter coefficient iteration model according to the current error value and the historical error value under the condition that the error value is not less than a preset error limit;

[0014] The processing module is used to perform denoising processing on the signal data using the updated filter coefficients.

[0015] The embodiments of this specification have at least the following advantages or beneficial effects:

[0016] The fuze denoising method based on adaptive filtering can effectively enhance the convergence speed of the above model and make the steady-state error performance better by constructing the above filter coefficient iteration model by variable constant parameters. Moreover, the above method can also introduce the historical error values ​​at different times before the current error value, so that the above method has the advantage of strong practicality and can be applied to a variety of different application scenarios. In addition, the convergence speed and steady-state error performance of the above method can be further improved by introducing the historical error values ​​at different times before the current moment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present specification and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 A flowchart of a fuze denoising method based on adaptive filtering provided in this specification;

[0019] Figure 2 This is a schematic diagram of a fuze denoising method based on adaptive filtering provided in this manual, using matlab to simulate a frequency modulated continuous wave transmission signal;

[0020] Figure 3 It is a schematic diagram of the spectrum of the FM continuous wave transmission signal of the MATLAB simulation of the fuse denoising method based on adaptive filtering provided in this manual;

[0021] Figure 4 This is a schematic diagram of a fuze denoising method based on adaptive filtering provided in this manual, using matlab to simulate a frequency modulated continuous wave echo signal;

[0022] Figure 5 It is a schematic diagram of the spectrum of the FM continuous wave echo signal of the MATLAB simulation of the fuse denoising method based on adaptive filtering provided in this manual;

[0023] Figure 6 This is a schematic diagram of the signal after FM continuous wave echo filtering of the fuze denoising method based on adaptive filtering provided in this manual in Matlab simulation;

[0024] Figure 7 It is a schematic diagram of the spectrum of the signal after the FM continuous wave echo filtering of the fuze denoising method based on adaptive filtering provided in this manual in Matlab simulation;

[0025] Figure 8 This is the error square variation diagram of the adaptive process of the fuze denoising method based on adaptive filtering provided in this manual in matlab simulation;

[0026] Fig. 9 It is the step-size iteration diagram of the matlab simulation adaptive process of the fuse denoising method based on adaptive filtering provided in this manual;

[0027] Fig.10 A schematic diagram of a fuze denoising system based on adaptive filtering provided in this specification.

[0028] Fig.11 It is a block diagram of a FIR type adaptive digital filter of a fuse denoising method based on adaptive filtering provided in this specification;

[0029] Fig.12 It is a schematic diagram of the working process of the adaptive filter provided in this manual;

[0030] Fig.13 It is a schematic diagram of a step-size update function module of a fuze denoising method based on adaptive filtering provided in this specification;

[0031] Fig.14 It is a schematic diagram of the step length update process of the fuze denoising method based on adaptive filtering provided in this specification;

[0032] Fig.15 It is a schematic diagram of the timing control of the step update module of the fuse denoising method based on adaptive filtering provided in this specification. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solutions and advantages of the embodiments of this specification more clear, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are part of the embodiments of this specification, not all of the embodiments. Generally, the components of the embodiments of this specification described and shown in the drawings here can be arranged and designed in various different configurations.

[0034] Please refer to Figures 1 to 9 An embodiment of the present specification provides a method for denoising a fuse based on adaptive filtering, which mainly includes:

[0035] Step 102: According to the variable step size algorithm model, a filter coefficient iteration model is constructed by variableizing the constant parameters;

[0036] Step 104, calculating the error value between the filter input signal data and the filter output signal data;

[0037] Step 106: Under the condition that the error value is not less than a preset error limit, the filter coefficients of the filter are iteratively updated through the filter coefficient iteration model according to the current error value and the historical error value;

[0038] Step 108: De-noising the signal data using the updated filter coefficients.

[0039] In this embodiment, the variable step size algorithm model is:

[0040] μ(n)=karctan(α|e(n)| m )

[0041] Among them, n is the time of signal data, μ(n) is the step size, α, k, m are constants, and e(n) is the current error value.

[0042] That is, the filter coefficient iteration model is constructed by variableizing the constant α.

[0043] In this embodiment, the historical error value is the error value calculated before the current error value, that is, the error value at different times before the current time.

[0044] Specifically, by constructing the above filter coefficient iteration model in a variable constant parameter manner, the convergence speed of the above model can be effectively enhanced, and the steady-state error performance can be improved. Moreover, the above method can also introduce historical error values ​​at different times before the current error value, and can not highlight the influence of different past times on the update of the filter coefficient at the current time, so that the above method has the advantage of strong practicality, and can be applicable to a variety of different application scenarios. In addition, by introducing the historical error values ​​at different times before the current time, the convergence speed and steady-state error performance of the above method can be further improved.

[0045] In this embodiment, after step 104, the following steps are further included:

[0046] Step 110: Under the condition that the error value is less than a preset error limit, the signal data is denoised using the current filter coefficient of the filter.

[0047] In this embodiment, the filter coefficient of the above filter is iteratively updated by judging the size relationship between the error value and the preset error limit. If the calculated current error value is smaller than the preset error limit, it indicates that the filter coefficient of the current filter can meet the requirements of signal data denoising.

[0048] In this embodiment, the above-mentioned preset error limit can be determined according to the actual needs of the current application scenario, and can also be determined according to empirical values.

[0049] In this embodiment, the effect of adaptively adjusting the filter coefficient of the filter is achieved by comparing the error values.

[0050] In this embodiment, one implementation of step 102 is as follows:

[0051] Step 122: According to the variable step size algorithm model, a step size iteration model is determined by variableizing constant parameters;

[0052] Step 124: construct the filter coefficient iteration model according to the step size iteration model, the current filter coefficient of the filter and the error value of the signal data.

[0053] Specifically, the constant parameter α of the above variable step size algorithm model intelligent light is variable into a variable related to the current error value and the historical error value, which can consider the influence of earlier error values ​​on the current error value, further improve the convergence speed of the above filter coefficient iterative update, and reduce the steady-state error.

[0054] In this embodiment, one implementation of step 122 is as follows:

[0055] Step 132: Variable the constant parameter α to α(n), which is:

[0056]

[0057] Among them, p is the error ratio coefficient of the feedback control function, q is the feedback depth, e(n) is the error value, n is the nth acquisition moment of the signal data, the feedback control function error ratio coefficient p and the feedback depth q are working parameter data from the DSP chip or SOC chip, and ni is the nith acquisition moment of the signal data.

[0058] Specifically, the constant parameter α can be parameterized into a variable related to the feedback depth, the current error value and the historical error value in the above manner, which can further improve the convergence speed of the iterative update of the above filter coefficients and reduce the steady-state error.

[0059] In this embodiment, the feedback depth can also be variable and quantified instead of being fixed to 1. The feedback depth can be determined according to actual requirements (ie, the processing requirements of the radio fuse echo).

[0060] In this embodiment, the feedback depth q represents the time depth of the impact of the historical error value on the current step size μ(n). The larger the value, the more historical error values ​​participate in the feedback control, and the historical error values ​​at different times are weighted. The earlier error values ​​are weighted to decrease exponentially by 0.5, thereby introducing but not highlighting the impact of the earlier error values ​​on the current time. It also has the following advantages: the parameter variable module is highly practical and suitable for a variety of different application scenarios, and the convergence speed and steady-state error are further improved. Step 134: According to the variable α(n), determine the step size iteration model μ(n), that is:

[0061] μ(n)=k·arctan(α(n)·e(n) m )

[0062] Wherein, k is the step function amplitude factor, m is the feedback control function error order, and the step function amplitude factor k and the feedback control function error order m are working parameter data from a DSP chip or a SOC chip.

[0063] In this embodiment, the step-size iteration model constructed according to the parameterized variable α(n) satisfies the following convergence condition: 0≤μ(n)≤1 / λ max ,λ max is the maximum eigenvalue of the autocorrelation matrix of x(n), where x(n) is the input signal data of the above filter.

[0064] In this embodiment, x(n) is a sequence of input signal data.

[0065] In this embodiment, one implementation of step 124 is as follows:

[0066] Step 142: construct an initial filter coefficient iteration model according to the step size iteration model, the current filter coefficient of the filter and the error value of the signal data, that is:

[0067] W(n+1)=W(n)+2μ(n)e(n)x(n)

[0068] Wherein, W(n+1) is the filter coefficient after iterative update of the filter, W(n) is the current filter coefficient of the filter, and x(n) is the filter input signal data.

[0069] Step 144: normalize the step size iteration model according to the power of the filter input signal data to obtain the filter coefficient iteration model, which is:

[0070]

[0071] Among them, τ is a constant, x T(n) is the transposed matrix of x(n).

[0072] In this embodiment, the filter coefficients can be iteratively updated by the specific method of step 142, and the interference of gradient noise amplification on the input signal data can be effectively avoided by the above step 144. In detail, the power of the above input signal data is replaced by the time average value of the signal. Therefore, the above method can further accelerate the convergence.

[0073] In this embodiment, the constant τ can prevent the denominator of the above formula from approaching zero.

[0074] In this embodiment, before step 104, the following steps are also included:

[0075] Step 152, collecting signal data according to the acquired data collection points at the pulse rising edge;

[0076] Step 154, storing the collected signal data and the current filter coefficients of the filter in a memory;

[0077] Step 156, sequentially calling the collected signal data as input signal data of the filter;

[0078] Step 158: Call the current filter coefficients of the filter.

[0079] In this embodiment, the collected signal data can be stored in the memory RAM in the above manner, that is, cache input, and then the signal data and filter coefficients in the above memory are called through instructions or communication protocols (the initial filter coefficients can be initial values ​​or preset values), and after the call, the above filter coefficients are iteratively updated.

[0080] In this embodiment, the updated filter coefficients may also be stored in the above-mentioned memory.

[0081] In this embodiment, one implementation of step 104 is as follows:

[0082] Step 162: Obtain output signal data of the filter according to the input signal data of the filter and the current filter coefficient of the filter;

[0083] Step 164: Obtain the error value according to the input signal data of the filter and the output signal data of the filter.

[0084] In this embodiment, the input signal data of the filter is x(n), the output data of the filter is y(n), and the expected signal data is d(n), then:

[0085] y(n)=W T (n)x(n)

[0086] e(n)=d(n)-y(n)

[0087] Among them, W T (n) is the current filter coefficient of the filter.

[0088] In this embodiment, the input signal data may be generated by simulating the superposition of an ideal signal and a noise signal, and in order to consider the stationary characteristics of the noise in the input signal data, the input signal data is used as the expected signal.

[0089] In this embodiment, the input signal data is filtered with updated filter coefficients, so that the useful part of the input signal data with strong correlation with the ideal signal is retained, and the noise part with weak correlation is filtered out, thereby achieving the effect of removing noise.

[0090] In this embodiment, in order to further demonstrate the effect of the above method, the following simulation example is used to illustrate in detail:

[0091] The simulation conditions are as follows: the working bandwidth of the linear frequency modulation continuous wave fuze is 1MHz; the signal time width is 200us; the sampling rate is 200MHz; the echo signal-to-noise ratio is 0dB; the step normalization parameter τ is 1, the filter order is 128, the step function amplitude factor k is 0.1, the feedback control function error order m is 2, the error ratio coefficient p is 2, and the feedback depth q is 2.

[0092] For detailed simulation results, please refer to Figures 2 to 9 :

[0093] according to Figures 2 to 5 It can be seen that the echo signal-to-noise ratio is 0dB, and it can be seen that the signal time domain waveform is almost "submerged" in the noise;

[0094] according to Figure 6 and Figure 7 It can be seen that the useful part of the time domain waveform of the filtered signal is obviously highlighted, the noise in the frequency domain is basically filtered out, and the frequency domain components are concentrated within the working bandwidth, which verifies the effectiveness of denoising;

[0095] according to Figure 8 It can be seen that as the number of iterations increases, the error square shows an overall downward trend, and the final steady-state error tends to 0, and converges after 118 iterations. The convergence speed is fast, which verifies the effectiveness of the filter adaptive process;

[0096] according to Fig. 9It can be seen that the overall change range of the step size is within 0.1, and the change trend is roughly the same as the trend of the square of the error value. When the square of the error value is large, the above method can accelerate the convergence speed with a large step size. When the square of the error value is small, the above method can reduce the steady-state error with a small step size, which verifies the effectiveness of the step size iteration model.

[0097] It can be seen that the above simulation results can verify the correctness, effectiveness and reliability of the above method.

[0098] Please refer to Figures 10 to 15 Another embodiment of the present specification provides a fuze denoising system based on adaptive filtering, which mainly includes:

[0099] A construction module 202 is used to construct a filter coefficient iteration model by variableizing constant parameters according to a variable step size algorithm model;

[0100] A calculation module 204 is used to calculate the error value between the filter input signal data and the output signal data;

[0101] An iteration module 206, configured to iteratively update the filter coefficients of the filter through the filter coefficient iteration model according to the current error value and the historical error value under the condition that the error value is not less than a preset error limit;

[0102] The processing module 208 is used to perform denoising processing on the signal data using the updated filter coefficients.

[0103] In this embodiment, the processing module 208 is used to perform denoising processing on the signal data using the current filter coefficient of the filter under the condition that the error value is less than a preset error limit.

[0104] It can be seen that by constructing the above filter coefficient iteration model in a variable constant parameter manner, the convergence speed of the above model can be effectively enhanced, and the steady-state error performance can be improved. Moreover, the above method can also introduce historical error values ​​at different times before the current error value, so that the above method has the advantage of strong practicality and can be applied to a variety of different application scenarios. In addition, by introducing historical error values ​​at different times before the current moment, the convergence speed and steady-state error performance of the above method can be further improved.

[0105] In this embodiment, the above modules can be implemented based on FPGA. The FPGA chip includes IP modules such as multipliers and CORDIC, which has the advantages of high real-time performance and fast processing speed, and is conducive to actual engineering.

[0106] In this embodiment, the construction module 202 is used to determine the step size iteration model by variableizing the constant parameter according to the variable step size algorithm model; and to construct the filter coefficient iteration model according to the step size iteration model, the current filter coefficient of the filter and the error value of the signal data. The constant parameter α is variableized to α(n), that is: Wherein, p is the error ratio coefficient of the feedback control function, q is the feedback depth, e(n) is the error value, n is the nth acquisition time of the signal data, the error ratio coefficient p of the feedback control function and the feedback depth q are the working parameter data from the DSP chip or the SOC chip, and ni is the nith acquisition time of the signal data. According to the variable α(n), the step size iteration model μ(n) is determined, that is: μ(n) = k·arctan(α(n)·e(n) m ), where k is the step function amplitude factor, m is the feedback control function error order, and the step function amplitude factor k and the feedback control function error order m are working parameter data from a DSP chip or a SOC chip. According to the step iteration model, the current filter coefficient of the filter and the error value of the signal data, an initial filter coefficient iteration model is constructed, namely: W(n+1)=W(n)+2μ(n)e(n)x(n), where W(n+1) is the filter coefficient after iterative update of the filter, W(n) is the current filter coefficient of the filter, and x(n) is the filter input signal data. The step iteration model is normalized according to the power of the filter input signal data to obtain the filter coefficient iteration model, namely: Among them, τ is a constant, x T (n) is the transposed matrix of x(n). The effect of adaptively adjusting the filter coefficient of the filter is achieved by comparing the above error values. Moreover, by parameterizing the constant parameter α of the above variable step size algorithm model into a variable related to the current error value and the historical error value, the influence of the earlier error value on the current error value can be considered, and the convergence speed of the iterative update of the above filter coefficient can be further improved, and the steady-state error can be reduced. In addition, by the above method, the constant parameter α variable can be parameterized into a variable related to the feedback depth, the current error value and the historical error value, which can further improve the convergence speed of the iterative update of the above filter coefficient and reduce the steady-state error.

[0107] In this embodiment, the system also includes a storage module 210, which is used to collect signal data according to the acquired data collection points at the rising edge of the pulse; store the collected signal data and the current filter coefficient of the filter in the memory; sequentially call the collected signal data as the input signal data of the filter; and call the current filter coefficient of the filter.

[0108] In this embodiment, the calculation module 204 is used to obtain the output signal data of the filter according to the input signal data of the filter and the current filter coefficient of the filter; and obtain the error value according to the input signal data of the filter and the output signal data of the filter. That is, the calculation module can synchronously call the signal data and filter coefficients stored in the RAM by the storage module, perform multiplication and accumulation operations on the signal data and the filter coefficients, and use the multiplication and accumulation results as the output of the filter, that is, the output signal data, and use the difference between the expected signal data and the output signal data as the calculated error value. Before calculating the error value, the output signal data and the expected signal data are truncated to the same signal bit width and meet the same fixed-point quantization criteria.

[0109] In this embodiment, since the error value only involves even-number power operations in the step size update process, the square of the error value is used as the output.

[0110] In this embodiment, the above-mentioned iteration module 206 can receive the working parameter data from the DSP processing chip or the SOC processing chip through the communication interface EMIF, which can be specifically the preset error limit, the filter order, the step function amplitude factor, the feedback control function error ratio coefficient, the feedback control function error order, and the feedback depth. The above-mentioned storage module 210 can receive the number of data acquisition points in a single cycle from the DSP processing chip or the SOC processing chip. After comparing the error value and the error limit, the above-mentioned iteration module can iteratively update the filter parameters according to the above-mentioned working parameter data under the condition that the error value is not less than the above-mentioned preset error limit, and output the end mark to the top layer of the above-mentioned iteration module after the iteration ends, until receiving the next re-iteration instruction from the DSP or SOC.

[0111] In this embodiment, the iterative update function is encapsulated into a generalized module, as shown in the following example: Fig.13 As shown in the figure, the update process of the above generalized module is as follows Fig.14 k, m, q, and p are the working parameter data from DSP or SOC, which are processed across clock domains using a delay device (i.e., Delay) and reduce the register fan-out. The square of the error value is e(n) 2 After the input, it enters the RAM cache. The parameter q determines the RAM read enable and the length of the read address. The larger the q, the more past errors involved in the feedback control. The read past errors and e(n) 2Enter the divider IP core, the division result is truncated and enters the multiplier IP core together with the delay register of parameter p. The multiplication result is truncated to the low (i-1) bit (i.e. the exponent with the previous error weight of 0.5) and then accumulated to obtain α(n). m is an even number, which is a necessary condition for the filter error to converge. m ≥ 2, so the parameter m is truncated to the low log2m bit as the number of times e(n)2 is multiplied by itself to complete e(n). m The cordic IP core completes the calculation of the inverse tangent function. The timing control in the iteration module is mainly completed by the input error valid flag, such as Fig.15 As shown, the error valid flag controls the generation of RAM read enable and the input valid flag of the divider IP core. After the IP core automatically calculates, it generates the division output valid flag, and then completes the delay of the output valid flag according to the fixed delay of the multiplier IP core and related logic, and controls the truncation, accumulator, and cordic IP core, finally realizing timing control and pipeline processing.

[0112] In this embodiment, the above-mentioned iterative module also has the advantages of high real-time performance, fast processing speed, high accuracy, generalization, parameterization and easy transplantation, and is suitable for different platforms and different systems. Moreover, compared with traditional methods, its noise removal effect is significant.

[0113] Based on the same invention, another embodiment of the present specification further provides a computer-readable storage medium, wherein the computer-readable storage medium stores one or more programs, and when the one or more programs are executed by an electronic device including multiple application programs, the electronic device executes Figure 1 The corresponding embodiment provides a fuse denoising method based on adaptive filtering.

[0114] In this specification, each embodiment is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0115] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0116] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, this specification may be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, this specification may be in the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0117] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that has the function specified in a box.

[0118] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0119] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0120] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0121] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0122] The above is only an embodiment of the present application and is not intended to limit this specification. For those skilled in the art, this specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this specification should be included in the scope of the claims of this specification.

Claims

1. A method for denoising a fuze based on adaptive filtering, characterized in that: include: According to the variable step size algorithm model, the filter coefficient iteration model is constructed by variableizing the constant parameters; Calculate the error value between the filter input signal data and the output signal data; Under the condition that the error value is not less than a preset error limit, according to the current error value and the historical error value, the filter coefficient of the filter is iteratively updated through the filter coefficient iteration model; The signal data is subjected to denoising processing using the updated filter coefficients.

2. The method for fuze denoising based on adaptive filtering according to claim 1, characterized in that: After calculating the error value between the filter input signal data and the output signal data, the method further comprises: Under the condition that the error value is less than a preset error limit, the signal data is denoised using the current filter coefficient of the filter.

3. The method for fuze denoising based on adaptive filtering according to claim 1, characterized in that: The variable step size algorithm model is used to construct a filter coefficient iteration model by variable constant parameters, including: According to the variable step size algorithm model, the step size iteration model is determined by variable constant parameters; The filter coefficient iteration model is constructed according to the step size iteration model, the current filter coefficient of the filter and the error value of the signal data.

4. The method for fuze denoising based on adaptive filtering according to claim 3 is characterized in that: The step size iteration model is determined by variably varying constant parameters according to the variable step size algorithm model, including: The constant parameter α is converted to α(n), which is: Among them, p is the error ratio coefficient of the feedback control function, q is the feedback depth, e(n) is the error value, n is the nth acquisition moment of the signal data, the feedback control function error ratio coefficient p and the feedback depth q are working parameter data from the DSP chip or SOC chip, and ni is the nith acquisition moment of the signal data.

5. The method for fuze denoising based on adaptive filtering according to claim 4 is characterized in that: After the constant parameter α is variable to α(n), the following steps are included: According to the variable α(n), the step size iteration model μ(n) is determined, which is: μ(n)=k·arctan(α(n)·e(n) m ) Wherein, k is the step function amplitude factor, m is the feedback control function error order, and the step function amplitude factor k and the feedback control function error order m are working parameter data from a DSP chip or a SOC chip.

6. The method for fuze denoising based on adaptive filtering according to claim 5 is characterized in that: The step of constructing the filter coefficient iteration model according to the step size iteration model, the current filter coefficient of the filter and the error value of the signal data comprises: According to the step size iteration model, the current filter coefficient of the filter and the error value of the signal data, an initial filter coefficient iteration model is constructed, that is: W(n+1)=W(n)+2μ(n)e(n)x(n) Wherein, W(n+1) is the filter coefficient after iterative update of the filter, W(n) is the current filter coefficient of the filter, and x(n) is the filter input signal data.

7. The method for fuze denoising based on adaptive filtering according to claim 6 is characterized in that: After constructing the initial filter coefficient iteration model according to the step size iteration model, the current filter coefficient of the filter and the error value of the signal data, the method further includes: The step size iteration model is normalized according to the power of the filter input signal data to obtain the filter coefficient iteration model, which is: Among them, τ is a constant, x T (n) is the transposed matrix of x(n).

8. The method for fuze denoising based on adaptive filtering according to claim 1, characterized in that: Before calculating the error value between the filter input signal data and the output signal data, the method includes: At the rising edge of the pulse, signal data is collected according to the number of acquired data collection points; storing the collected signal data and the current filter coefficients of the filter in a memory; sequentially calling the collected signal data as input signal data of the filter; Recalls the current filter coefficients for the filter.

9. The method for fuze denoising based on adaptive filtering according to claim 8, characterized in that: The calculating the error value between the filter input signal data and the output signal data includes: Obtaining output signal data of the filter according to input signal data of the filter and current filter coefficients of the filter; The error value is obtained according to the input signal data of the filter and the output signal data of the filter.

10. A fuze denoising system based on adaptive filtering, characterized in that: include: A construction module is used to construct a filter coefficient iteration model by variableizing constant parameters according to a variable step size algorithm model; A calculation module, used for calculating the error value between the filter input signal data and the output signal data; An iteration module, configured to iteratively update the filter coefficients of the filter through the filter coefficient iteration model according to the current error value and the historical error value under the condition that the error value is not less than a preset error limit; The processing module is used to perform denoising processing on the signal data using the updated filter coefficients.

Images