Pulse signal adaptive filtering method and system based on multi-band signal entropy

Abstract

The invention provides a pulse signal adaptive filtering method and system based on multi-band signal entropy. The method comprises the following steps: receiving a plurality of satellite communication signals of different frequency bands, and extracting a phase offset, an amplitude value and a frequency value; and generating a multi-dimensional feature group of each frequency band, and extracting the signal-to-noise ratio and the frequency variation of the corresponding frequency band from the multi-dimensional feature group. And based on the signal-to-noise ratios and the frequency variations of all the frequency bands, dynamically adjusting filtering parameters through a preset three-stage tuning structure, and filtering to obtain a filtered signal. Transmitting the filtered signal to a federated learning framework, and calculating a multi-band signal entropy value by combining the multi-dimensional feature group; adaptive filtering parameters are optimized based on entropy values, global aggregation is carried out after differential privacy processing, and an aggregation feature graph group is generated; and through a channel and a space attention mechanism, effective information of a high-entropy frequency band is enhanced, noise of a low-entropy frequency band is suppressed, and pulse signal adaptive filtering based on multi-frequency-band signal entropy is completed. According to the technical scheme provided by the invention, the accuracy of adaptive filtering of the pulse signal is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of adaptive filtering, in particular to a pulse signal adaptive filtering method and system based on multi-band signal entropy values. BACKGROUND

[0002] In scenarios such as satellite communication and radar detection that rely on pulse signal transmission, multi-band pulse signals often face interference from complex electromagnetic environments, and effective signals in high-entropy bands and noise interference in low-entropy bands are often mixed and superimposed. To ensure the stability and accuracy of signal transmission, adaptive filtering technology is needed to dynamically distinguish between effective information and noise, accurately preserve high-entropy band signals, and efficiently suppress low-entropy band noise, while adapting to the dynamic changes in signal characteristics of different bands, which puts strict requirements on the real-time performance and relevance of the filtering method.

[0003] Currently, for adaptive filtering of multi-band pulse signals, a common solution is to construct a filtering model based on a single feature (such as signal-to-noise ratio), and to adjust the filtering parameters uniformly for signals in each band using a pre-set fixed threshold. This solution first extracts a single feature value of the signal, and then adjusts the filtering parameters linearly based on the comparison result of the feature value and the threshold, to achieve filtering of signals in different bands.

[0004] However, this solution has obvious defects: relying on a single feature cannot fully reflect the complex characteristics of multi-band pulse signals, and the differences in characteristics between high-entropy bands and low-entropy bands are difficult to accurately capture; the fixed threshold parameter adjustment method lacks dynamic adaptability and is difficult to match real-time changes in signal characteristics, which can easily lead to excessive filtering of effective information in high-entropy bands or excessive noise residue in low-entropy bands, ultimately affecting the transmission quality of pulse signals and the accuracy of subsequent processing. SUMMARY

[0005] The present application provides a pulse signal adaptive filtering method and system based on multi-band signal entropy values to solve the problems of incomplete feature reflection, lack of dynamic adaptation, and poor filtering effect in the prior art.

[0006] In a first aspect, the embodiments of the present application provide a pulse signal adaptive filtering method based on multi-band signal entropy values, comprising: receiving satellite communication signals in multiple different bands, and extracting phase offsets, amplitude values, and frequency values in the satellite communication signals; based on the phase offsets, amplitude values, and frequency values, generating a plurality of multi-dimensional feature groups corresponding to each band, and extracting signal-to-noise ratios and frequency changes of the corresponding band from each of the multi-dimensional feature groups; Based on the signal-to-noise ratio and the frequency variation of all frequency bands, dynamically adjusting the filter characteristic parameters through a preset three-stage tuning structure, and filtering the satellite communication signals of the different frequency bands based on the dynamically adjusted filter characteristic parameters to obtain filtered signals; Transmitting the filtered signal to a federated learning framework, and calculating a multi-band signal entropy value based on the federated learning framework and the multidimensional feature set; Based on the multi-band signal entropy values, optimizing the adaptive filtering parameters, and globally aggregating the optimized adaptive filtering parameters through differential privacy protection rules to generate an aggregated feature map group; Based on the channel and spatial attention mechanism, the aggregated feature map group is weighted to enhance the effective information of the high entropy frequency band and suppress the noise interference of the low entropy frequency band, thereby completing the adaptive filtering of the pulse signal based on the entropy value of the multi-band signal.

[0007] Optionally, the dynamically adjusting the filtering characteristic parameters based on the signal-to-noise ratio and the frequency variation of all frequency bands through a preset three-level tuning structure includes: Based on the signal-to-noise ratios of all frequency bands, filter bandwidth adjustment parameters of corresponding frequency bands are calculated according to a preset inverse proportional relationship to construct a first-level tuning structure; Based on the frequency changes of all frequency bands, center frequency adjustment parameters of corresponding frequency bands are calculated according to a preset proportional correspondence to construct a second-level tuning structure; Integrating the signal-to-noise ratios and the frequency variations of all frequency bands, and obtaining a signal comprehensive evaluation parameter based on the integration result through a preset weighted calculation method; Based on the comprehensive signal evaluation parameters, the signal gain adjustment parameters of the corresponding frequency band are determined according to a preset corresponding relationship to construct a third-level tuning structure; Integrating the first-stage tuning structure, the second-stage tuning structure, and the third-stage tuning structure to obtain a preset three-stage tuning structure; Based on the three-stage tuning structure, the filter bandwidth adjustment parameter, the center frequency adjustment parameter and the signal gain adjustment parameter are mapped to the filter characteristics of the corresponding frequency band to dynamically adjust the filter characteristic parameters to obtain the dynamically adjusted filter characteristic parameters.

[0008] Optionally, based on the three-stage tuning structure, mapping the filter bandwidth adjustment parameter, the center frequency adjustment parameter, and the signal gain adjustment parameter to the filter characteristics of the corresponding frequency band to dynamically adjust the filter characteristic parameters to obtain the dynamically adjusted filter characteristic parameters includes: Determine a basic parameter set for each frequency band, wherein the basic parameter set includes a basic bandwidth parameter, a basic center frequency parameter, and a basic gain parameter; Based on a preset bandwidth mapping rule, the filter bandwidth adjustment parameter in the three-stage tuning structure is combined with the basic bandwidth parameter of the corresponding frequency band to obtain a dynamic bandwidth parameter; Based on a preset frequency mapping rule, the center frequency adjustment parameter in the three-level tuning structure is combined with the basic center frequency parameter of the corresponding frequency band to obtain a dynamic center frequency parameter; Based on a preset gain mapping rule, the signal gain adjustment parameter in the three-stage tuning structure is combined with the basic gain parameter of the corresponding frequency band to obtain a dynamic gain parameter; The dynamic bandwidth parameter, the dynamic center frequency parameter, and the dynamic gain parameter are mapped to the filter characteristic of the corresponding frequency band to generate dynamically adjusted filter characteristic parameters of the corresponding frequency band.

[0009] Optionally, transmitting the filtered signal to a federated learning framework, and calculating a multi-band signal entropy value based on the federated learning framework and the multidimensional feature group includes: Based on a preset frequency band division standard, the filtered signal is divided into a plurality of sub-signals, and the plurality of sub-signals are respectively transmitted to nodes corresponding to each frequency band in the federated learning framework, so that after each node receives the sub-signal of the corresponding frequency band, the sub-signal is associated with the multidimensional feature group, and based on the association result, characteristic parameters are calculated on the associated data of the sub-signal and the multidimensional feature group according to a preset local calculation rule to obtain signal characteristic distribution parameters of the corresponding frequency band; Based on the federated learning framework and the preset cross-node collaboration rules, the signal feature distribution parameters of the corresponding frequency bands of each node are integrated to form a global feature distribution parameter; The global characteristic distribution parameters are calculated according to a preset entropy value calculation rule to obtain the multi-band signal entropy value.

[0010] Optionally, the step of performing characteristic parameter calculation on the correlation data between the sub-signal and the multi-dimensional feature group based on the correlation result and according to a preset local calculation rule to obtain the signal characteristic distribution parameters of the corresponding frequency band includes: Comparing the phase offset of the associated data in the multidimensional feature group with a preset reference phase value to generate a phase difference value sequence; Extracting a maximum difference and a minimum difference from the phase difference sequence, and calculating a phase fluctuation range parameter based on the maximum difference and the minimum difference; Based on the time domain distribution statistical rule in the preset local calculation rule, the amplitude value of the correlation data is calculated to obtain the amplitude change rate parameter; Calculating the difference between the frequency value in the associated data and a preset reference frequency value to obtain a frequency offset parameter; Constructing a characteristic parameter set according to the phase fluctuation range parameter, the amplitude change rate parameter, and the frequency offset parameter; A distribution characteristic analysis is performed on the characteristic parameter set, and based on the result of the distribution characteristic analysis, a signal characteristic distribution parameter of the corresponding frequency band is generated.

[0011] Optionally, optimizing the adaptive filtering parameters based on the multi-band signal entropy values, and globally aggregating the optimized adaptive filtering parameters through differential privacy protection rules to generate an aggregated feature map group includes: Determining the adjustment weight of the adaptive filtering parameters according to the entropy value of the multi-band signal; Based on the adjustment weights, the initial adaptive filtering parameters are optimized step by step to obtain optimized adaptive filtering parameters; Based on the differential privacy protection rule, the privacy perturbation value that meets the preset threshold condition is superimposed on the optimized adaptive filtering parameters to obtain the target adaptive filtering parameters; Based on a preset parameter fusion ratio, the target adaptive filtering parameters are input into an aggregation node in the federated learning framework and then globally integrated to generate global parameters; The global parameters are reorganized according to preset feature dimensions to generate an aggregated feature map group.

[0012] Optionally, the channel- and space-based attention mechanism weights the aggregated feature map group to enhance the effective information of the high-entropy frequency band and suppress the noise interference of the low-entropy frequency band, thereby completing the pulse signal adaptive filtering based on the entropy value of the multi-band signal, including: Extracting initial channel features and initial spatial features corresponding to each frequency band from the aggregated feature map group; Determining a first weighting coefficient and a second weighting coefficient for the initial channel feature and the initial spatial feature, respectively, based on the multi-band signal entropy value of each frequency band and an attention mechanism; Based on the first weighting coefficient, the initial channel feature is enhanced or attenuated to obtain a target channel feature; Based on the second weighting coefficient, performing regional weight allocation on the initial spatial feature to obtain a target spatial feature; Superimposing and fusing the target channel features with the target spatial features to generate a weighted aggregate feature map group; The weighted aggregated feature map group is used to enhance the effective information of the high entropy value frequency band and suppress the noise interference of the low entropy value frequency band, so as to complete the pulse signal adaptive filtering based on the entropy value of the multi-band signal.

[0013] In a second aspect, an embodiment of the present application provides a pulse signal adaptive filtering system based on multi-band signal entropy values, including: receive a plurality of satellite communication signals of different frequency bands, extract a phase offset, an amplitude value and a frequency value in the satellite communication signals; a first generation module, configured to generate a plurality of multi-dimensional feature groups corresponding to respective frequency bands based on the phase offset, the amplitude value and the frequency value, and extract a signal-to-noise ratio and a frequency variation of a corresponding frequency band from each of the multi-dimensional feature groups; an adjustment module, configured to dynamically adjust filter characteristic parameters through a preset three-level tuning structure based on the signal-to-noise ratios and the frequency variations of all the frequency bands, and perform filter processing on the satellite communication signals of the different frequency bands based on the dynamically adjusted filter characteristic parameters to obtain filter signals; a transmission module, configured to transmit the filter signals to a federated learning framework, and calculate multi-band signal entropy values based on the federated learning framework and the multi-dimensional feature groups; a second generation module, configured to optimize adaptive filter parameters based on the multi-band signal entropy values, and perform global aggregation on the optimized adaptive filter parameters through a differential privacy protection rule to generate an aggregated feature map group; a weighting module, configured to perform weighting on the aggregated feature map group based on a channel and spatial attention mechanism, so as to enhance effective information of a high-entropy-value frequency band and suppress noise interference of a low-entropy-value frequency band, and complete pulse signal adaptive filtering based on multi-band signal entropy values.

[0014] In a third aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component, and implement the pulse signal adaptive filtering method based on multi-band signal entropy values according to any one of the first aspect.

[0015] In a fourth aspect, an embodiment of the present application provides a computer storage medium, which stores a computer program; when the computer program is executed by a computer, a pulse signal adaptive filtering method based on multi-band signal entropy values is implemented.

[0016] The present application provides a pulse signal adaptive filtering method based on multi-band signal entropy value, including: receiving satellite communication signals of multiple frequency bands, extracting phase offset, amplitude value and frequency value in the satellite communication signals; generating multidimensional feature groups corresponding to each frequency band based on the phase offset, amplitude value and frequency value, extracting the signal-to-noise ratio and frequency variation of the corresponding frequency band from each multidimensional feature group; based on the signal-to-noise ratio and frequency variation of all frequency bands, dynamically adjusting the filtering characteristic parameters through a preset three-level tuning structure, and adjusting the filtering characteristic parameters of different frequency bands based on the dynamically adjusted filtering characteristic parameters. The satellite communication signal is filtered to obtain a filtered signal; the filtered signal is transmitted to the federated learning framework, and the entropy value of the multi-band signal is calculated based on the federated learning framework and the multi-dimensional feature group; based on the entropy value of the multi-band signal, the adaptive filtering parameters are optimized, and the optimized adaptive filtering parameters are globally aggregated through the differential privacy protection rule to generate an aggregated feature map group; based on the channel and spatial attention mechanism, the aggregated feature map group is weighted to enhance the effective information of the high entropy value frequency band and suppress the noise interference of the low entropy value frequency band, thereby completing the adaptive filtering of the pulse signal based on the entropy value of the multi-band signal.

[0017] The present application has the following advantages: by receiving satellite communication signals of multiple different frequency bands and extracting phase offset, amplitude and frequency values, the basic characteristic parameters of the signal can be obtained, providing raw data for subsequent characteristic analysis; by generating a multidimensional feature group based on the phase offset, amplitude and frequency values ​​and extracting the signal-to-noise ratio and frequency variation, the basic characteristics can be converted into key indicators reflecting the signal quality and dynamic characteristics, providing a basis for adjusting the filtering parameters; by dynamically adjusting the filtering characteristic parameters based on the signal-to-noise ratio and frequency variation of all frequency bands using a preset three-level tuning structure and obtaining a filtered signal, the filtering process can be initially adapted to the characteristic differences of the signals in each frequency band; The filtered signal is transmitted to the federated learning framework, and the entropy value of the multi-band signal is calculated in combination with the multi-dimensional feature group. The distributed framework can be used to obtain quantitative indicators reflecting the complexity of the signal in each frequency band; by optimizing the adaptive filtering parameters based on the entropy value of the multi-band signal, the aggregated feature graph group is generated by global aggregation through differential privacy protection rules, which can protect the privacy of the parameters while obtaining the global features of the fusion of multi-node information; the aggregated feature graph group is weighted by the channel-based and spatial attention mechanism, the effective information of the high-entropy frequency band is enhanced and the noise interference of the low-entropy frequency band is suppressed to complete the filtering, which can enable the effective signal to be targetedly enhanced and the noise to be accurately suppressed, thereby realizing adaptive filtering of the pulse signal.

[0018] Further, based on the signal-to-noise ratio and the frequency variation of all frequency bands, a first-level tuning structure containing a filter bandwidth adjustment parameter, a second-level tuning structure containing a center frequency adjustment parameter, and a third-level tuning structure containing a signal gain adjustment parameter are constructed, respectively. After integrating the three-level tuning structures, the above adjustment parameters are mapped to the filter characteristics of the corresponding frequency band in combination with the basic parameter set of each frequency band, to obtain the dynamically adjusted filter characteristic parameters. The application enables multi-dimensional dynamic adaptation of the filter characteristic parameters according to the signal-to-noise ratio and the frequency variation of each frequency band, thereby improving the matching degree of the filter characteristics and the signal characteristics of each frequency band.

[0019] These aspects or other aspects of the application will be more apparent from the following description of the embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0021] Figure 1 A flowchart of a pulse signal adaptive filtering method based on multi-band signal entropy values provided by an embodiment of the application; Figure 2 A structural schematic diagram of a pulse signal adaptive filtering system based on multi-band signal entropy values provided by an embodiment of the application; Figure 3 A structural schematic diagram of a computing device provided by an embodiment of the application. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the technical solutions of the application, the technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application.

[0023] In some processes described in the specification and claims of the application and the above description, a plurality of operations appearing in a specific order are included, but it should be clearly understood that these operations can be executed in the order appearing in this document or in parallel. The serial numbers of the operations, such as 11, 12, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes can include more or fewer operations, and the operations can be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this document are used to distinguish different messages, devices, modules, etc., and do not represent the order of sequence, nor do "first" and "second" represent different types.

[0024] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0025] In order to solve the problems of poor pulse signal filtering caused by incomplete feature reflection, lack of dynamic adaptation, and poor filtering effect in the existing technology, the embodiment of the present application provides a pulse signal adaptive filtering method based on multi-band signal entropy value. The method adopts the following concept: accurate filtering of multi-band satellite pulse signals is achieved through step-by-step processing: from obtaining basic signal information to converting it into key indicators, and then performing preliminary filtering through a dynamic tuning structure, combining the federated learning framework to calculate signal complexity, integrating multi-source features while protecting privacy, and finally using the attention mechanism to specifically enhance effective information and suppress noise. The synergistic effect of each step enables the filtering process to adapt to changes in signal characteristics, taking into account both information retention and noise suppression, thereby improving the stability and accuracy of pulse signal transmission.

[0026] Figure 1 A flowchart of a pulse signal adaptive filtering method based on multi-band signal entropy values ​​provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the method includes: S11. Receive multiple satellite communication signals in different frequency bands, and extract phase offset, amplitude, and frequency values ​​from the satellite communication signals.

[0027] Among them, multiple satellite communication signals in different frequency bands refer to signals transmitted by satellites in different frequency ranges (signals in different frequency bands fluctuate at different speeds); the phase offset is the degree of deviation of the signal fluctuation position relative to the reference position; the amplitude value reflects the strength of the signal; the frequency value is the number of signal fluctuations per second; the results generated in this step are the phase offset, amplitude value and frequency value of each frequency band.

[0028] In an embodiment of the present application, a receiving device acquires satellite signals of different frequency bands, and a measurement tool is used to sequentially determine the phase offset, amplitude, and frequency values ​​of each signal. For example, after receiving signals from frequency band A (fluctuation range of 10 million to 20 million times per second) and frequency band B (20 million to 30 million times per second), the measurement shows that frequency band A deviates from the reference position by 25 degrees (phase offset), has a strength of 4 units (amplitude), and fluctuates by 15 million times per second (frequency). The corresponding values ​​for frequency band B are obtained similarly.

[0029] S12. Generate a multidimensional feature group corresponding to each frequency band based on the phase offset, amplitude value, and frequency value, and extract the signal-to-noise ratio and frequency variation of the corresponding frequency band from each multidimensional feature group.

[0030] Among them, the multidimensional feature group is a collection of phase offset, amplitude value, and frequency value in the same frequency band; the signal-to-noise ratio is the ratio of effective signal to noise; the larger the ratio, the clearer the effective signal; the frequency change is the change in signal frequency over a period of time; the results generated in this step are the signal-to-noise ratio and frequency change of each frequency band.

[0031] In this embodiment of the present application, for each frequency band, the phase offset, amplitude, and frequency values ​​extracted by S11 are integrated into a multidimensional feature set, and the signal-to-noise ratio and frequency change are calculated from the feature set. For example, the multidimensional feature set of frequency band A includes 25 degrees, 4 units, and 15 million times / second. When calculating the signal-to-noise ratio, the effective signal amplitude is 4 units and the noise amplitude is 0.2 units, and the signal-to-noise ratio is 4 ÷ 0.2 = 20. When calculating the frequency change, the frequency is 15 million times / second in the previous second and 15.01 million times / second in the next second, and the change is 10,000 times / second.

[0032] S13. Based on the signal-to-noise ratio and frequency variation of all frequency bands, the filter characteristic parameters are dynamically adjusted through a preset three-stage tuning structure. Based on the dynamically adjusted filter characteristic parameters, satellite communication signals of different frequency bands are filtered to obtain filtered signals.

[0033] The three-stage tuning structure is a combination of three parts for adjusting filter settings; filter characteristic parameters refer to settings that affect the filtering effect (such as filter range, center position, and amplification level); dynamic adjustment refers to the change of parameters as signal characteristics change; the filtered signal is the signal after filtering; the result generated by this step is the filtered signal of each frequency band.

[0034] In this embodiment, a three-stage tuning structure is constructed based on the signal-to-noise ratio and frequency variation. The first stage calculates the filter bandwidth adjustment parameter based on the signal-to-noise ratio. For example, a signal-to-noise ratio of 20 corresponds to a bandwidth adjustment parameter of 2 million times / second. The second stage calculates the center frequency adjustment parameter based on the frequency variation. For example, a variation of 20,000 times / second corresponds to an adjustment parameter of 200,000 times / second. The third stage uses weighted calculation to integrate the signal-to-noise ratio and frequency variation to obtain a comprehensive evaluation parameter, which determines the signal gain adjustment parameter. For example, a comprehensive evaluation parameter of 12.4 corresponds to a gain of 1.2 times. After integrating the three-stage structure, the filter characteristic parameters are adjusted, and then the signal is filtered.

[0035] S14. Transmit the filtered signal to the federated learning framework, and calculate the multi-band signal entropy value based on the federated learning framework and the multi-dimensional feature group.

[0036] Among them, the federated learning framework is a system in which multiple devices collaborate to process data; the multi-band signal entropy value is a numerical value that reflects the complexity of the signal. The larger the value, the richer the information may be; the result generated in this step is the signal entropy value of each frequency band.

[0037] In an embodiment of the present application, the filtered signal is transmitted to the federated learning framework, and the entropy value is calculated in combination with the multidimensional feature group. The formula is H=-Σ(p_i×log(p_i)), where H is the entropy value (, and p_i is the feature occurrence ratio.

[0038] S15. Based on the entropy value of the multi-band signal, optimize the adaptive filtering parameters, and globally aggregate the optimized adaptive filtering parameters through the differential privacy protection rule to generate an aggregated feature map group.

[0039] Among them, adaptive filtering parameters are filtering settings that can automatically adjust with signal changes; differential privacy protection rules protect data privacy by adding random values; global aggregation is the process of integrating the processing results of multiple devices; the aggregated feature map group is a collection of images containing multi-band features after integration; the result generated by this step is an aggregated feature map group.

[0040] In an embodiment of the present application, the adaptive filtering parameters are optimized based on the entropy value, and the higher the entropy value, the finer the parameters; random values ​​within a preset range are added according to the rules, such as the optimization parameter 20 plus 0.3 becomes 20.3; the parameters are transmitted to the aggregation node, and the multi-node parameters are integrated according to a preset ratio (such as each node accounts for 50%); and finally, they are organized into an aggregate feature map group.

[0041] S16. Based on the channel and space attention mechanism, the aggregated feature map group is weighted to enhance the effective information of the high entropy frequency band and suppress the noise interference of the low entropy frequency band, completing the adaptive filtering of the pulse signal based on the entropy value of the multi-band signal.

[0042] Among them, the channel and space attention mechanism is a method of focusing on the important parts of the signal. The channel refers to the source of the signal, and the space refers to the position of the signal. Weighting is to assign different degrees of importance to different parts. The high entropy frequency band refers to the frequency band with a large entropy value and may contain more effective information. The low entropy frequency band refers to the frequency band with a small entropy value and may contain more noise. The result generated in this step is the completion of the adaptive filtering of the pulse signal.

[0043] In an embodiment of the present application, the channel characteristics and spatial characteristics of each frequency band are extracted from the aggregated feature map group; the weighting coefficient is determined based on the entropy value (the higher the entropy value, the higher the coefficient, such as an entropy value of 1.20 corresponds to 0.8, and 0.85 corresponds to 0.2); the characteristics of the high-entropy frequency band are enhanced and the characteristics of the low-entropy frequency band are weakened according to the coefficient, and finally the filtering is completed.

[0044] For example, when processing multi-band satellite pulse signals, signals from bands A, B, and C are first received, and the phase offset, amplitude, and frequency values ​​for each band are extracted. A multidimensional feature set is then generated for each band, and the signal-to-noise ratio (SNR) for bands A, B, and C (20, 15, and 25) and frequency variation (10,000, 20,000, and 5,000 times / second) are calculated. Based on these metrics, a three-level tuning structure is constructed: bandwidth adjustment parameters are calculated based on the SNR, center frequency adjustment parameters are calculated based on the frequency variation, and the gain adjustment parameters are determined by integrating the two comprehensive evaluation parameters. After integration, the filter characteristic parameters are adjusted and filtering is performed. The filtered signal is then fed into the federated learning framework, where, combined with the feature set, entropy values ​​for each band (1.03, 0.85, and 1.20) are calculated using H = -Σ(pi × log(pi)). Filter parameters are optimized based on the entropy values, and random values ​​are added (for example, 20 plus 0.3 for band A is 20.3). The filter parameters are then transmitted to the aggregation node and integrated with the parameters of other nodes to generate an aggregated feature map. Finally, the weighting coefficients (C=0.8, B=0.2) are set according to the entropy value to enhance the characteristics of the high-entropy frequency band and suppress the characteristics of the low-entropy frequency band to complete the adaptive filtering.

[0045] By executing S11-S16, the present embodiment achieves precise filtering of multi-band satellite pulse signals through a step-by-step process: from acquiring basic signal information to converting it into key indicators, then performing preliminary filtering through a dynamic tuning structure, combining it with a federated learning framework to calculate signal complexity, integrating multi-source features while protecting privacy, and finally using an attention mechanism to specifically enhance effective information and suppress noise. The synergistic effect of these steps enables the filtering process to adapt to changes in signal characteristics, balancing information preservation and noise suppression, thereby improving the stability and accuracy of pulse signal transmission.

[0046] In a possible embodiment, S13, based on the signal-to-noise ratio and frequency variation of all frequency bands, dynamically adjusting the filter characteristic parameters through a preset three-level tuning structure includes: Step 131 : Based on the signal-to-noise ratios of all frequency bands, filter bandwidth adjustment parameters of corresponding frequency bands are calculated according to a preset inverse proportional relationship to construct a first-level tuning structure.

[0047] Among them, the signal-to-noise ratio is the ratio of the effective signal to the noise. The larger the ratio, the clearer the signal. The filter bandwidth adjustment parameter is used to adjust the frequency range of the signal allowed to pass. The first-level tuning structure is composed of the filter bandwidth adjustment parameters of each frequency band and is used to adjust the filter range. This step generates the first-level tuning structure.

[0048] In an embodiment of the present application, the signal-to-noise ratios of all frequency bands are collected, and the parameters of each frequency band are calculated according to a preset inverse proportional relationship (the larger the signal-to-noise ratio, the smaller the filter bandwidth adjustment parameter). For example, the preset filter bandwidth adjustment parameter = 1000 ÷ signal-to-noise ratio, the signal-to-noise ratios of bands A, B, and C are 20, 15, and 25, respectively. Then, the parameter of band A = 1000 ÷ 20 = 50, band B = 1000 ÷ 15 ≈ 67, and band C = 1000 ÷ 25 = 40. These parameters are then used to construct the first-level tuning structure.

[0049] Step 132: Based on the frequency variations of all frequency bands, center frequency adjustment parameters of corresponding frequency bands are calculated according to a preset proportional correspondence to construct a second-level tuning structure.

[0050] Among them, the frequency variation is the degree of change of the signal frequency over a period of time; the center frequency adjustment parameter is used to adjust the center position of the filtering range; the second-level tuning structure is composed of the center frequency adjustment parameters of each frequency band and is used to adjust the center position of the filter; this step generates the second-level tuning structure.

[0051] In an embodiment of the present application, the frequency changes of all frequency bands are collected, and the parameters of each frequency band are calculated according to a preset proportional correspondence (the larger the frequency change, the larger the center frequency adjustment parameter). For example, the preset center frequency adjustment parameter = frequency change × 20, and the frequency changes of bands A, B, and C are 10,000 times / second, 20,000 times / second, and 5,000 times / second, respectively. Then, the parameter of band A = 1×20=20, band B = 2×20=40, and band C = 0.5×20=10, and these parameters are then used to construct a second-level tuning structure.

[0052] Step 133: Integrate the signal-to-noise ratios and frequency variations of all frequency bands, and obtain a comprehensive signal evaluation parameter based on the integration result through a preset weighted calculation method.

[0053] Among them, integration is the process of combining the signal-to-noise ratio and the frequency change; the weighted calculation method is the method of calculating after assigning different weights to the two; the signal comprehensive evaluation parameter is a value that comprehensively reflects the two and is used to comprehensively evaluate the signal; this step generates the comprehensive evaluation parameters of the signal in each frequency band.

[0054] In this embodiment of the present application, the weights of the signal-to-noise ratio and the frequency variation are determined (e.g., the signal-to-noise ratio is 60% and the frequency variation is 40%), and a comprehensive evaluation parameter is calculated for each frequency band. For example, if the signal-to-noise ratio of frequency band A is 20 and the frequency variation is 10,000 times / second, the parameter = 20×60%+1×40%=12+0.4=12.4; frequency band B = 15×60%+2×40%=9+0.8=9.8; and frequency band C = 25×60%+0.5×40%=15+0.2=15.2.

[0055] Step 134 : Based on the comprehensive signal evaluation parameters, the signal gain adjustment parameters of the corresponding frequency band are determined according to a preset corresponding relationship to construct a third-level tuning structure.

[0056] The signal gain adjustment parameter is used to adjust the signal amplification degree; the third-level tuning structure is composed of the signal gain adjustment parameters of each frequency band and is used to adjust the amplification degree; this step generates the third-level tuning structure.

[0057] In an embodiment of the present application, a correspondence between the comprehensive evaluation parameter and the gain adjustment parameter is preset (e.g., for every increase of 1 in the parameter, the gain increases by 0.1, and the initial value is 1), and the parameters of each frequency band are calculated. For example, the comprehensive evaluation parameter of frequency band A is 12.4, the gain parameter = 1 + 12.4 × 0.1 = 2.24; frequency band B = 1 + 9.8 × 0.1 = 1.98; frequency band C = 1 + 15.2 × 0.1 = 2.52, and then these parameters are used to construct the third-level tuning structure.

[0058] Step 135 : Integrate the first-level tuning structure, the second-level tuning structure, and the third-level tuning structure to obtain a preset three-level tuning structure.

[0059] Among them, the first, second and third level tuning structures adjust the filtering range, center position and amplification degree respectively; the preset three-level tuning structure is a complete adjustment system combining the three; this step generates the preset three-level tuning structure.

[0060] In an embodiment of the present application, the first, second, and third-level tuning parameters of each frequency band are combined in the order of filtering range, center position, and amplification degree. For example, the combination of frequency band A is (50, 20, 2.24), frequency band B is (67, 40, 1.98), and frequency band C is (40, 10, 2.52). Integrating these combinations forms a three-level tuning structure.

[0061] Step 136 : Based on the three-stage tuning structure, the filter bandwidth adjustment parameter, the center frequency adjustment parameter, and the signal gain adjustment parameter are mapped to the filter characteristics of the corresponding frequency band to dynamically adjust the filter characteristic parameters to obtain dynamically adjusted filter characteristic parameters.

[0062] Among them, mapping is the process of applying adjustment parameters to filter characteristics; filter characteristics include range, center position, amplification degree, etc.; the dynamically adjusted filter characteristic parameters are parameters that are adjusted to better fit the signal characteristics; this step generates the dynamically adjusted filter characteristic parameters for each frequency band.

[0063] In an embodiment of the present application, the parameters in the three-stage tuning structure are respectively applied to the corresponding filtering characteristics. For example, the original filtering range of the A band is 1 million to 2 million times / second, and after applying the bandwidth parameter 50, it becomes 1 million to 2.5 million times / second; the original center position is 1.5 million times / second, and after applying the center parameter 20, it becomes 1.7 million times / second; the original amplification factor is 1.0, and after applying the gain parameter 2.24, it becomes 2.24, and the adjusted filtering characteristic parameters are obtained.

[0064] For example, when a system processes satellite signals in bands A, B, and C, the signal-to-noise ratios (SNRs) for bands A, B, and C are known to be 20, 15, and 25, respectively. Using the filter bandwidth adjustment parameter = 1000 ÷ SNR, the filter bandwidth adjustment parameters for each band are 50, 67, and 40, respectively. These parameters are integrated to form the first-level tuning structure. The frequency variations for the three bands are known to be 10,000, 20,000, and 5,000 times / second, respectively. Using the center frequency adjustment parameter = frequency variation × 20, the values ​​are 20, 40, and 10, respectively. These parameters are integrated to form the second-level tuning structure. Using a weight of 60% for SNR and 40% for frequency variation, the signal comprehensive evaluation parameters for each band are calculated: 20 × 60% + 1 × 40% = 12.4 for band A, 15 × 60% + 2 × 40% = 9.8 for band B, and 25 × 60% + 0. 5×40%=15.2; calculated according to the signal gain adjustment parameter = 1 + comprehensive evaluation parameter × 0.1, the gain adjustment parameters of each frequency band are 2.24, 1.98, and 2.52, and these parameters are integrated to form a third-level tuning structure; the first, second, and third-level tuning parameters of the three frequency bands are respectively combined into (50, 20, 2.24), (67, 40, 1.98), and (40, 10, 2.52), and these combinations are integrated to form a preset three-level tuning structure; finally, these parameters are mapped to the filtering characteristics of the corresponding frequency bands. For example, the filtering range of frequency band A is adjusted from the original 1-2 million times / second to 1-2.5 million times / second, the center position is adjusted from 1.5 million times / second to 1.7 million times / second, and the amplification factor is adjusted from 1.0 to 2.24. Similarly, the filtering characteristic parameters of frequency bands B and C after dynamic adjustment are obtained.

[0065] By executing steps 131 to 136, the embodiment of the present application achieves dynamic adjustment of the filter characteristic parameters by constructing and integrating the tuning structure in three levels. From adjusting the filter range based on the signal-to-noise ratio, to adjusting the center position based on the frequency change, to adjusting the amplification level after comprehensive evaluation, each link is closely connected, so that the filter characteristics can fully fit the signal clarity, dynamic changes and overall performance. Ultimately, the filter characteristic parameters obtained through mapping are more targeted and can better adapt to the characteristics of signals in different frequency bands, improving the adaptability and accuracy of the filtering effect.

[0066] In one possible embodiment, step 136, based on the three-stage tuning structure, mapping the filter bandwidth adjustment parameter, the center frequency adjustment parameter, and the signal gain adjustment parameter to the filter characteristics of the corresponding frequency band to dynamically adjust the filter characteristic parameters to obtain the dynamically adjusted filter characteristic parameters, includes: a1. Determine the basic parameter set for each frequency band. The basic parameter set includes basic bandwidth parameters, basic center frequency parameters, and basic gain parameters.

[0067] The basic parameter set is the initial setting set for filtering in each frequency band, including the basic bandwidth parameter (the size of the signal frequency range initially allowed to pass), the basic center frequency parameter (the center position of the initial filtering range), and the basic gain parameter (the initial signal amplification factor). This step generates the basic parameter set for each frequency band.

[0068] In an embodiment of the present application, the basic bandwidth, basic center frequency and basic gain parameters of each frequency band are determined according to the signal characteristics of each frequency band and conventional filtering requirements. For example, the basic bandwidth of frequency band A is 1 million times / second (allowing signals within the range of 1 million times / second to pass), the basic center frequency is 1.5 million times / second (the center of the filtering range is 1.5 million times / second), and the basic gain is 1.0 times (no amplification); the basic bandwidth of frequency band B is 1.5 million times / second, the basic center frequency is 2.5 million times / second, and the basic gain is 1.0 times; the basic bandwidth of frequency band C is 2 million times / second, the basic center frequency is 3.5 million times / second, and the basic gain is 1.0 times. These parameters are sorted out to form the basic parameter set of each frequency band.

[0069] a2. Based on a preset bandwidth mapping rule, the filter bandwidth adjustment parameter in the three-level tuning structure is combined with the basic bandwidth parameter of the corresponding frequency band to obtain a dynamic bandwidth parameter.

[0070] The preset bandwidth mapping rule is a rule for combining the filter bandwidth adjustment parameter with the basic bandwidth parameter. The dynamic bandwidth parameter is the new signal frequency passing range obtained after the combination. This step generates the dynamic bandwidth parameter of each frequency band.

[0071] In an embodiment of the present application, a preset bandwidth mapping rule is clearly defined (such as dynamic bandwidth parameter = basic bandwidth parameter + filter bandwidth adjustment parameter), and the filter bandwidth adjustment parameter in the three-level tuning structure is combined with the basic bandwidth parameter of the corresponding frequency band according to this rule. For example, the basic bandwidth of frequency band A is 1 million times / second, the filter bandwidth adjustment parameter is 500,000 times / second, and the dynamic bandwidth = 1 million + 500,000 = 1.5 million times / second; the basic bandwidth of frequency band B is 1.5 million times / second, the adjustment parameter is 670,000 times / second, and the dynamic bandwidth = 1.5 million + 670,000 = 2.17 million times / second; the basic bandwidth of frequency band C is 2 million times / second, the adjustment parameter is 400,000 times / second, and the dynamic bandwidth = 2 million + 400,000 = 2.4 million times / second.

[0072] a3. Based on a preset frequency mapping rule, the center frequency adjustment parameter in the three-level tuning structure is combined with the basic center frequency parameter of the corresponding frequency band to obtain a dynamic center frequency parameter.

[0073] The preset frequency mapping rule is a rule for combining the center frequency adjustment parameter with the basic center frequency parameter. The dynamic center frequency parameter is the new center position of the filtering range obtained after the combination. This step generates the dynamic center frequency parameter of each frequency band.

[0074] In an embodiment of the present application, a preset frequency mapping rule is clearly defined (such as dynamic center frequency parameter = basic center frequency parameter + center frequency adjustment parameter), and the center frequency adjustment parameter in the three-level tuning structure is combined with the basic center frequency parameter of the corresponding frequency band according to this rule. For example, the basic center frequency of frequency band A is 1.5 million times / second, the center frequency adjustment parameter is 200,000 times / second, and the dynamic center frequency = 1.5 million + 200,000 = 1.7 million times / second; the basic center frequency of frequency band B is 2.5 million times / second, the adjustment parameter is 400,000 times / second, and the dynamic center frequency = 2.5 million + 400,000 = 2.9 million times / second; the basic center frequency of frequency band C is 3.5 million times / second, the adjustment parameter is 100,000 times / second, and the dynamic center frequency = 3.5 million + 100,000 = 3.6 million times / second.

[0075] a4. Based on a preset gain mapping rule, the signal gain adjustment parameter in the three-stage tuning structure is combined with the basic gain parameter of the corresponding frequency band to obtain a dynamic gain parameter.

[0076] The preset gain mapping rule is a rule for combining the signal gain adjustment parameter with the basic gain parameter. The dynamic gain parameter is a new signal amplification factor obtained after the combination. This step generates the dynamic gain parameter of each frequency band.

[0077] In the embodiment of the present application, a preset gain mapping rule is clearly defined (such as dynamic gain parameter = basic gain parameter × signal gain adjustment parameter), and the signal gain adjustment parameter in the three-stage tuning structure is combined with the basic gain parameter of the corresponding frequency band according to this rule. For example, the basic gain of frequency band A is 1.0 times, the signal gain adjustment parameter is 2.24, and the dynamic gain = 1.0×2.24=2.24 times; the basic gain of frequency band B is 1.0 times, the adjustment parameter is 1.98, and the dynamic gain = 1.0×1.98=1.98 times; the basic gain of frequency band C is 1.0 times, the adjustment parameter is 2.52, and the dynamic gain = 1.0×2.52=2.52 times.

[0078] a5. Mapping the dynamic bandwidth parameter, the dynamic center frequency parameter, and the dynamic gain parameter to the filter characteristic of the corresponding frequency band to generate the dynamically adjusted filter characteristic parameters of the corresponding frequency band.

[0079] Among them, the dynamic bandwidth parameter, dynamic center frequency parameter, and dynamic gain parameter are the adjusted filter range size, center position, and amplification factor, respectively. Mapping them to the filter characteristics means applying these parameters to the actual filter settings. The dynamically adjusted filter characteristic parameters are the final filter settings after integration. This step generates the dynamically adjusted filter characteristic parameters for each frequency band.

[0080] In an embodiment of the present application, the correspondence between the dynamic parameters and the filtering characteristics is determined (dynamic bandwidth corresponds to range size, dynamic center frequency corresponds to center position, and dynamic gain corresponds to amplification factor), and the three are integrated into the filtering settings of each frequency band. For example, the A band is centered on a dynamic center frequency of 1.7 million times / second, a dynamic bandwidth of 1.5 million times / second (range of 1.7 million ± 750,000 times / second), and an amplification of 2.24 times; the B band is centered on 2.9 million times / second, a bandwidth of 2.17 million times / second (range of 2.9 million ± 1.085 million times / second), and an amplification of 1.98 times; the C band is centered on 3.6 million times / second, a bandwidth of 2.4 million times / second (range of 3.6 million ± 1.2 million times / second), and an amplification of 2.52 times, forming the final filtering characteristic parameters.

[0081] For example, when a system processes satellite signals in bands A, B, and C, it first determines the basic parameter sets for each band: Band A has a basic bandwidth of 1 million times per second, a basic center frequency of 1.5 million times per second, and a basic gain of 1.0; Band B has a basic bandwidth of 1.5 million times per second, a basic center frequency of 2.5 million times per second, and a basic gain of 1.0; and Band C has a basic bandwidth of 2 million times per second, a basic center frequency of 3.5 million times per second, and a basic gain of 1.0. Then, according to the bandwidth mapping rule (dynamic bandwidth = basic bandwidth + filter bandwidth adjustment parameter), combined with the adjustment parameters of 500,000, 670,000, and 400,000 times per second in the three-level tuning structure, the dynamic bandwidths for Band A are calculated to be 1.5 million times per second (1 million + 500,000) times per second (1.5 million times per second), 2.17 million times per second (1.5 million + 670,000) times per second (2.17 million times per second) for Band B, and 2.4 million times per second (2 million + 400,000) times per second (2.4 million times per second) for Band C. Then, according to the frequency mapping rule (dynamic center frequency = basic center frequency + center frequency adjustment parameter), combined with the adjustment parameters of 200,000, 400,000, and 100,000 times / second, the dynamic center frequencies for Band A are 1.5 million + 200,000 = 1.7 million times / second, for Band B 2.5 million + 400,000 = 2.9 million times / second, and for Band C 3.5 million + 100,000 = 3.6 million times / second. Then, according to the gain mapping rule (dynamic gain = basic gain × signal gain adjustment parameter), combined with the adjustment parameters of 2.24, 1.98, and 2.52, the dynamic gains for Band A are 1.0 × 2.24 = 2.24, for Band B 1.0 × 1.98 = 1.98, and for Band C 1.0 × 2.52 = 2.52. Finally, these parameters are mapped to the filtering characteristics. The parameters of the A band are centered at 1.7 million times / second, with a bandwidth of 1.5 million times / second and an amplification of 2.24 times. The parameters of the B band are centered at 2.9 million times / second, with a bandwidth of 2.17 million times / second and an amplification of 1.98 times. The parameters of the C band are centered at 3.6 million times / second, with a bandwidth of 2.4 million times / second and an amplification of 2.52 times.

[0082] By executing steps a1 to a5, the present embodiment of the present invention achieves a complete transformation of filtering characteristics from basic settings to dynamic adaptation, from determining the initial filtering parameters to adjusting the parameters in combination with signal characteristics, and then integrating them into the final filtering settings. Each step uses clear mapping rules to rationally combine the basic parameters with the signal characteristic adjustment parameters, so that the filtering range, center position, and amplification factor can fully fit the signal characteristics of different frequency bands. The resulting filtering parameters are more targeted, can better adapt to signal changes, and improve the accuracy and adaptability of the filtering process.

[0083] In a possible embodiment, S14, transmitting the filtered signal to a federated learning framework, and calculating the multi-band signal entropy value based on the federated learning framework and the multi-dimensional feature group, includes: Step 141: Based on a preset frequency band division standard, the filtered signal is divided into multiple sub-signals, and the multiple sub-signals are respectively transmitted to the nodes corresponding to each frequency band in the federated learning framework, so that after each node receives the sub-signal of the corresponding frequency band, the sub-signal is associated with the multidimensional feature group, and based on the association result, according to the preset local calculation rules, the characteristic parameters of the associated data between the sub-signal and the multidimensional feature group are calculated to obtain the signal characteristic distribution parameters of the corresponding frequency band.

[0084] Among them, the preset frequency band division standard is a regulation for splitting the filtered signal into different frequency ranges; the filtered signal is the signal after preliminary processing; the sub-signal is the small signal corresponding to each frequency band after splitting; the federated learning framework is a system in which multiple processing units work together; the node is the processing unit corresponding to each frequency band in the system; the multidimensional feature group is a collection of information such as signal phase, amplitude, frequency, etc.; association is to correspond the sub-signal with the multidimensional feature group information of the same frequency band; the local calculation rule is a method for each node to independently calculate the feature parameters; the associated data is the associated sub-signal and multidimensional feature group information; the feature parameter calculation is to calculate the parameters that describe the signal characteristics; the signal feature distribution parameter is the parameter that reflects the distribution of the feature in the signal; this step generates the signal feature distribution parameters for each frequency band.

[0085] In an embodiment of the present application, the filtered signal is first split into three sub-signals A, B, and C according to a preset frequency band division standard (e.g., 10-20 million times / second is frequency band A, 20-30 million times / second is frequency band B, and 30-40 million times / second is frequency band C); the sub-signals are then transmitted to the nodes of the corresponding frequency bands in the federated learning framework (A to node 1, B to node 2, and C to node 3); after each node receives the sub-signal, it is associated with the multidimensional feature group of the corresponding frequency band (e.g., frequency band A contains information such as 25 degrees and 4 units), and then the associated data is calculated according to the local calculation rules (e.g., the proportion of the number of occurrences of the statistical feature in the three intervals) to obtain the signal feature distribution parameters. For example, node 1 counts the feature of sub-signal A 30 times in the first interval, 50 times in the second interval, and 20 times in the third interval (a total of 100 times), and the distribution parameters are 30, 50, and 20.

[0086] Step 142: Based on the federated learning framework and the preset cross-node collaboration rules, the signal feature distribution parameters of the corresponding frequency bands of each node are integrated to form global feature distribution parameters.

[0087] Among them, the cross-node collaboration rules are the regulations for integrating the parameters of each node in the federated learning framework; the signal feature distribution parameters are the feature distribution parameters calculated by each node; the global feature distribution parameters are the parameters that reflect the overall feature distribution after integration; this step generates the global feature distribution parameters.

[0088] In an embodiment of the present application, the cross-node collaboration rules are first determined (such as allocating weights according to the importance of the frequency band: A accounts for 40%, B accounts for 30%, and C accounts for 30%); then the signal feature distribution parameters of each node are collected (such as node 1: 30, 50, 20; node 2: 20, 60, 20; node 3: 40, 40, 20); and integrated according to the rules, for example, the global parameter of the first interval = 30×40%+20×30%+40×30%=12+6+12=30, the second interval = 50×40%+60×30%+40×30%=20+18+12=50, the third interval = 20×40%+20×30%+20×30%=8+6+6=20, and the global feature distribution parameters of 30, 50, and 20 are obtained.

[0089] Step 143: Calculate the global characteristic distribution parameters according to the preset entropy value calculation rules to obtain the multi-band signal entropy value.

[0090] Among them, the entropy value calculation rule is a method for calculating the complexity of the signal; the global feature distribution parameter is a parameter that reflects the overall feature distribution; the multi-band signal entropy value is a numerical value that reflects the complexity of the multi-band signal; this step generates the multi-band signal entropy value.

[0091] In the embodiment of the present application, the entropy calculation rule is first determined. Specifically, the proportion of each interval in the global feature distribution parameter is first calculated, and then the product of each proportion and the logarithm corresponding to the proportion is calculated respectively. Then, these products are added together, and finally the inverse of the sum is taken to obtain the entropy value.

[0092] For example, when a system processes filtered signals in the three frequency bands A, B, and C, it first splits the filtered signals into three sub-signals A, B, and C according to the preset standard (10-20 million times / second is A, 20-30 million times / second is B, and 30-40 million times / second is C), and transmits them to nodes 1, 2, and 3 of the federated learning framework respectively; each node associates the sub-signals with the multidimensional feature group of the corresponding frequency band, and calculates according to the local rule (the proportion of the occurrence of statistical features in the three intervals), and obtains the distribution parameters of frequency band A as 30, 50, and 20, frequency band B as 20, 60, and 20, and frequency band C as 40, 40, and 20; then according to the cross-node rule (A accounts for 40%, B accounts for 30%, and C accounts for 30%). The global parameter of the first interval is 30×0.4+20×0.3+40×0.3=30, the second interval is 50×0.4+60×0.3+40×0.3=50, and the third interval is 20×0.4+20×0.3+20×0.3=20, forming global parameters of 30, 50, and 20. Finally, the parameters are converted to 0.3, 0.5, and 0.2 according to the entropy formula, and the calculation result is (H=-(0.3×log0.3+0.5×log0.5+0.2×log0.2)=1.0295), and the entropy value of the multi-band signal is obtained.

[0093] By executing steps 141 through 143, the present embodiment achieves precise calculation of the characteristics of each frequency band by splitting the signal by frequency band and processing it by the corresponding nodes. Local parameters are integrated using cross-node collaborative rules to form global parameters that reflect the overall signal characteristics. Finally, the multi-band signal entropy calculated based on the global parameters fully reflects the complexity of the signal. The close connection between the steps preserves the uniqueness of each frequency band while integrating global information, providing a reliable basis for subsequent optimization of filtering parameters and improving the rationality and effectiveness of the overall processing.

[0094] In one possible embodiment, step 141, based on the correlation result, calculates characteristic parameters of the correlation data between the sub-signals and the multi-dimensional feature group according to a preset local calculation rule to obtain signal characteristic distribution parameters of the corresponding frequency band, including: b1. Compare the phase offset of the associated data in the multidimensional feature group with a preset reference phase value to generate a phase difference value sequence.

[0095] Among them, the phase offset of the associated data in the multidimensional feature group is a series of numerical values ​​in the signal after association that reflect the phase deviation from the reference position; the preset reference phase value is a standard phase value pre-set as a reference; the phase difference value sequence is a sequence of difference values ​​obtained by comparing the phase offset with the reference phase value. This step generates the phase difference value sequence.

[0096] In an embodiment of the present application, a series of phase offset values ​​are first extracted from the associated data. For example, the phase offsets of the A frequency band are 25 degrees, 28 degrees, 23 degrees, and 30 degrees, respectively. Each value is then compared with a preset reference phase value (for example, 0 degrees), and the reference phase value is subtracted from the phase offset to obtain a difference. That is, 25 degrees minus 0 degrees equals 25 degrees, 28 degrees minus 0 degrees equals 28 degrees, 23 degrees minus 0 degrees equals 23 degrees, and 30 degrees minus 0 degrees equals 30 degrees. Finally, these differences are arranged in order to form a phase difference sequence of 25 degrees, 28 degrees, 23 degrees, and 30 degrees.

[0097] b2. Extract the maximum difference and the minimum difference from the phase difference sequence, and calculate the phase fluctuation range parameter based on the maximum difference and the minimum difference.

[0098] Among them, the phase difference value sequence is a series of phase difference values ​​generated in step b1; the maximum difference is the largest value in the sequence; the minimum difference is the smallest value in the sequence; the phase fluctuation range parameter is the value reflecting the overall phase fluctuation amplitude obtained by subtracting the minimum difference from the maximum difference; this step generates the phase fluctuation range parameter.

[0099] In an embodiment of the present application, the maximum and minimum values ​​are first found from the phase difference sequence. For example, the phase difference sequence of the A frequency band in b1 is 25 degrees, 28 degrees, 23 degrees, and 30 degrees, where the maximum difference is 30 degrees and the minimum difference is 23 degrees. Then, the maximum difference is subtracted from the minimum difference, that is, 30 degrees minus 23 degrees equals 7 degrees, to obtain the phase fluctuation range parameter.

[0100] b3 calculates the amplitude value of the associated data based on the time domain distribution statistical rule in the preset local calculation rule to obtain the amplitude change rate parameter.

[0101] Among them, the time domain distribution statistical rule in the preset local calculation rule is a rule based on time-varying statistical data (such as calculating the change in unit time); the amplitude value of the correlation data is a series of numerical values ​​reflecting the strength of the signal after correlation; the amplitude change rate parameter is a numerical value reflecting the speed of change of the amplitude value in unit time; this step generates the amplitude change rate parameter.

[0102] In an embodiment of the present application, the statistical rules of time domain distribution are first determined, for example, the amplitude difference between two adjacent time points is calculated and then divided by the time interval; then the amplitude values ​​at different time points are extracted from the associated data, for example, the amplitude of the A frequency band at time t1 is 4 units, and at time t2 is 4.5 units, and the time interval is 1 second; according to the rules, the amplitude value of the latter time point is subtracted from the amplitude value of the previous time point, and then divided by the time interval, that is, (4.5 units minus 4 units) divided by 1 second is equal to 0.5 units / second, to obtain the amplitude change rate parameter.

[0103] b4. Calculate the difference between the frequency value in the associated data and the preset reference frequency value to obtain a frequency offset parameter.

[0104] The frequency value in the correlation data is a series of numerical values ​​reflecting the number of fluctuations per second in the correlated signal; the preset reference frequency value is a standard frequency value pre-set as a reference; the frequency offset parameter is the difference between the frequency value and the reference frequency value, reflecting the degree to which the frequency deviates from the reference; this step generates the frequency offset parameter.

[0105] In an embodiment of the present application, a frequency value is first extracted from the associated data. For example, the frequency value of frequency band A at a certain moment is 15.01 million times / second. Then, the frequency value is subtracted from a preset reference frequency value (such as 15 million times / second), that is, 15.01 million times / second minus 15 million times / second equals 10,000 times / second, to obtain a frequency offset parameter.

[0106] b5. Construct a characteristic parameter set based on the phase fluctuation range parameter, the amplitude change rate parameter, and the frequency offset parameter.

[0107] Among them, the phase fluctuation range parameter, the amplitude change rate parameter, and the frequency offset parameter are parameters that respectively reflect the phase fluctuation amplitude, the amplitude change speed, and the degree of frequency deviation; the characteristic parameter set is a set of data formed by integrating these three parameters, which is used to comprehensively describe the signal characteristics; this step generates the characteristic parameter set.

[0108] In the embodiment of the present application, the phase fluctuation range parameter obtained by b2, the amplitude change rate parameter obtained by b3, and the frequency offset parameter obtained by b4 are first collected. For example, the A frequency band is 7 degrees, 0.5 units / second, and 10,000 times / second respectively; then these three parameters are integrated in sequence to form a characteristic parameter set (7 degrees, 0.5 units / second, 10,000 times / second).

[0109] b6. Perform distribution characteristic analysis on the characteristic parameter set, and generate signal characteristic distribution parameters of the corresponding frequency band based on the results of the distribution characteristic analysis.

[0110] Among them, the characteristic parameter set is the comprehensive parameter group formed in step b5; the distribution characteristic analysis is to analyze the proportion of the number of times the values ​​of each parameter in the parameter set fall in different intervals; the result after the distribution characteristic analysis is the proportion data of each interval obtained by analysis; the signal characteristic distribution parameter is a numerical value reflecting the distribution of the characteristic parameters in different intervals; this step generates the signal characteristic distribution parameters of the corresponding frequency band.

[0111] In an embodiment of the present application, a method for analyzing the distribution characteristics is first determined, for example, the value range of each parameter is divided into three intervals, and the proportion of the number of times falling in each interval is counted; then each parameter in the characteristic parameter set is analyzed, for example, the phase fluctuation range of 7 degrees in the A band falls in the 5-10 degree interval (accounting for 100%), the amplitude change rate of 0.5 units / second falls in the 0-1 unit / second interval (accounting for 100%), and the frequency offset of 10,000 times / second falls in the 0-20,000 times / second interval (accounting for 100%); finally, the interval proportions of each parameter are integrated to generate the signal characteristic distribution parameters (100%, 100%, 100%).

[0112] For example, when a system processes the associated data of A frequency band, it first extracts the phase offset values 25 degrees, 28 degrees, 23 degrees and 30 degrees from the associated data, compares them with the reference phase value 0 degrees, calculates the difference values 25-0=25 degrees, 28-0=28 degrees, 23-0=23 degrees and 30-0=30 degrees, and generates a phase difference value sequence; then finds the maximum difference value 30 degrees and the minimum difference value 23 degrees from the sequence, calculates the phase fluctuation range parameter 30-23=7 degrees; extracts the amplitude value 4 units at t1 and the amplitude value 4.5 units at t2 (interval 1 second), calculates the amplitude change rate parameter (4.5-4) ÷ 1=0.5 units / second according to the rule; then extracts the frequency value 1501 million times / second, compares it with the reference frequency 1500 million times / second, and obtains the frequency offset parameter 1501 million-1500 million=10,000 times / second; integrates the three parameters into a feature parameter set (7 degrees, 0.5 units / second, 10,000 times / second); and finally determines that each parameter interval accounts for 100% according to the distribution analysis method, and generates the signal feature distribution parameter (100%, 100%, 100%).

[0113] By performing b1~b6, the embodiment of the present application quantifies the signal features from the three dimensions of phase, amplitude and frequency through step-by-step processing: first, the phase fluctuation range is obtained by comparing the phase with the reference value, reflecting the phase stability; then the amplitude change rate is analyzed in the time dimension, reflecting the signal strength dynamics; then the offset parameter is obtained by comparing the frequency with the reference value, reflecting the fluctuation stability; finally, these parameters are integrated and the distribution is analyzed to form comprehensive feature distribution parameters. The steps are closely connected, from local feature extraction to global distribution analysis, providing multi-dimensional and quantifiable basis for subsequent signal processing, and improving the comprehensiveness and accuracy of signal feature description.

[0114] In a possible embodiment, S15, based on the multi-frequency band signal entropy value, optimizes the adaptive filtering parameter, performs global aggregation on the optimized adaptive filtering parameter through the differential privacy protection rule, and generates an aggregated feature map group, including: Step 151, according to the size of the multi-frequency band signal entropy value, determine the adjustment weight of the adaptive filtering parameter.

[0115] Wherein, the multi-frequency band signal entropy value is a value reflecting the complexity of multiple frequency band signals, the larger the entropy value, the more information the signal may contain; the adaptive filtering parameter is a filtering setting parameter that can be automatically adjusted according to the signal feature; the adjustment weight is a value reflecting the importance of adjusting the filtering parameter of different frequency bands, the larger the weight, the higher the adjustment priority; this step generates the adjustment weight of the adaptive filtering parameter of each frequency band.

[0116] In an embodiment of the present application, the multi-band signal entropy value of each frequency band is first collected, for example, 1.03 for band A, 0.85 for band B, and 1.20 for band C; then the adjustment weight is determined according to the entropy value, and the total weight is 100% distributed proportionally. The total entropy value is 3.08, the weight of band C is 1.20÷3.08×100%≈39%, the weight of band A is 1.03÷3.08×100%≈33%, and the weight of band B is 0.85÷3.08×100%≈28%.

[0117] Step 152: Based on the adjustment weights, the initial adaptive filtering parameters are optimized step by step to obtain optimized adaptive filtering parameters.

[0118] Among them, the adjustment weight is the importance of adjustment of each frequency band determined in step 151; the initial adaptive filtering parameters are the unoptimized basic filtering parameters; the step-by-step optimization is to adjust the parameters from high to low according to the weight, and the adjustment range is large for high weights; the optimized adaptive filtering parameters are parameters that are more in line with the signal characteristics after adjustment; this step generates the optimized adaptive filtering parameters.

[0119] In the embodiment of the present application, the optimization order (C→A→B) is first determined; the initial parameters are set (A: 2.0, B: 1.8, C: 2.2); the adjustment amplitude is determined according to the weight (every 10% weight corresponds to an amplitude of 0.1), 39% of the C band corresponds to 0.4, and after optimization, 2.2+0.4=2.6; 33% of the A band corresponds to 0.3, and after optimization, 2.0+0.3=2.3; 28% of the B band corresponds to 0.3, and after optimization, 1.8+0.3=2.1.

[0120] Step 153: Based on the differential privacy protection rule, a privacy disturbance value that meets the preset threshold condition is superimposed on the optimized adaptive filtering parameter to obtain the target adaptive filtering parameter.

[0121] Among them, the differential privacy protection rule is a rule for protecting data by adding random values; the optimized adaptive filtering parameters are the adjusted parameters of step 152; the privacy disturbance value is a random value within the range of ±0.5; the target adaptive filtering parameter is the final parameter after superimposing the disturbance value; this step generates the target adaptive filtering parameter.

[0122] In the embodiment of the present application, the rules are first clarified (adding random values ​​within ±0.5); to optimize the parameters, the disturbance values ​​are superimposed, and the frequency band A is 2.3+0.2=2.5, the frequency band B is 2.1-0.3=1.8, and the frequency band C is 2.6+0.4=3.0.

[0123] Step 154: Based on the preset parameter fusion ratio, the target adaptive filtering parameters are input into the aggregation node in the federated learning framework and globally integrated to generate global parameters.

[0124] Among them, the preset parameter fusion ratio is the proportion of each node parameter in the global integration (such as 50% for each of the two nodes); the target adaptive filtering parameter is the final parameter of step 153; the aggregation node is the processing unit for integrating the parameters of each node; the global parameter is the parameter that reflects the overall situation after integration; this step generates the global parameter.

[0125] In the embodiment of the present application, the fusion ratio is first determined (50% for each of the two nodes); the parameters of node 1 (A: 2.5, B: 1.8, C: 3.0) and node 2 (A: 2.6, B: 1.7, C: 3.1) are input into the aggregation node; and integrated proportionally, frequency band A is 2.5×50%+2.6×50%=2.55, frequency band B is 1.8×50%+1.7×50%=1.75, and frequency band C is 3.0×50%+3.1×50%=3.05.

[0126] Step 155: reorganize the global parameters according to the preset feature dimensions to generate an aggregated feature map group.

[0127] Among them, the global parameters are the integrated parameters of step 154; the preset feature dimensions are the classification angles of the parameters (such as filtering range, center position, etc.); the aggregated feature graph group is a set of parameter graphs organized by dimension; this step generates the aggregated feature graph group.

[0128] In the embodiment of the present application, the feature dimensions (filter range, center position, and magnification degree) are first determined; the global parameters are reorganized by dimension, with frequency band A 2.55 corresponding to the filter range, frequency band B 1.75 corresponding to the center position, and frequency band C 3.05 corresponding to the magnification degree; and these are organized into charts to form an aggregated feature graph group.

[0129] For example, when a system processes signals in frequency bands A, B, and C, it first collects entropy values ​​for each band (A: 1.03, B: 0.85, C: 1.20), resulting in a total entropy of 3.08. Adjustment weights are then determined proportionally (C: 39%, A: 33%, B: 28%). Based on the weights, the initial parameters (A: 2.0, B: 1.8, C: 2.2) are optimized. After adjusting the C band by 0.4, it becomes 2.6; after adjusting A by 0.3, it becomes 2.3; and after adjusting B by 0.3, it becomes 2.1. According to differential privacy rules, the perturbation values ​​are added: A: 2.3 + 0.2 = 2.5, B: 2.1 - 0.3 = 1.8, and C: 2.6 + 0.4 = 3.0. The parameters of the two nodes (Node 1: A:2.5, B:1.8, C:3.0; Node 2: A:2.6, B:1.7, C:3.1) are combined at a ratio of 50% to obtain the global parameters A:2.55, B:1.75, C:3.05. Finally, the parameters are reorganized according to the filter range, center position, and magnification to generate an aggregated feature map group.

[0130] By executing steps 151 through 155, this embodiment of the present application forms a complete parameter processing system, from determining adjustment priorities based on signal complexity, to optimizing parameters by weight, to ensuring data security through privacy protection rules, and finally integrating multi-source parameters and organizing them by feature dimension. These closely linked steps ensure that the filtering parameters accurately adapt to signal characteristics while also taking data security into account. The resulting aggregated feature graphs are intuitive and clear, providing a comprehensive and reliable basis for subsequent filtering processing and improving the rationality and security of the overall processing.

[0131] In one possible embodiment, S16, based on the channel and spatial attention mechanism, weights the aggregated feature map group to enhance the effective information of the high entropy frequency band and suppress the noise interference of the low entropy frequency band, and completes the pulse signal adaptive filtering based on the entropy value of the multi-band signal, including: Step 161: extract the initial channel features and initial spatial features corresponding to each frequency band from the aggregated feature map group.

[0132] Among them, the aggregated feature graph group is a set of parameter graphs organized according to feature dimensions; the initial channel features are the original data extracted from the graph group, reflecting the characteristics of the signal in different transmission channels; the initial spatial features are the original data extracted from the graph group, reflecting the characteristics of the signal in spatial distribution; this step generates the initial channel features and initial spatial features corresponding to each frequency band.

[0133] In an embodiment of the present application, the charts corresponding to each frequency band in the aggregated feature graph group are first determined, for example, the filter range chart corresponding to frequency band A, the center position chart corresponding to frequency band B, and the magnification degree chart corresponding to frequency band C; then, data reflecting the channel characteristics are extracted from these charts as initial channel features, such as frequency band A is extracted to 5, frequency band B is 4, and frequency band C is 6; at the same time, data reflecting the spatial distribution is extracted as the initial spatial features, such as frequency band A is 3, frequency band B is 2, and frequency band C is 5.

[0134] Step 162: Based on the multi-band signal entropy value of each frequency band and the attention mechanism, determine a first weighting coefficient and a second weighting coefficient for the initial channel feature and the initial spatial feature, respectively.

[0135] Among them, the multi-band signal entropy value is a numerical value reflecting the complexity of the signal in each frequency band; the attention mechanism is a method of highlighting important features and weakening secondary features; the first weighting coefficient is a numerical value used to adjust the importance of the initial channel features; the second weighting coefficient is a numerical value used to adjust the importance of the initial spatial features; this step generates the first and second weighting coefficients.

[0136] In an embodiment of the present application, the multi-band signal entropy values ​​of each frequency band are first collected, for example, 1.03 for band A, 0.85 for band B, and 1.20 for band C; then, combined with the attention mechanism, the coefficient is determined according to the principle that the larger the entropy value, the higher the coefficient, and the total coefficient sum is 1. The total entropy value is calculated to be 1.03+0.85+1.20=3.08, and the first weighted coefficient is 1.20÷3.08≈0.4 for band C, 1.03÷3.08≈0.3 for band A, and 0.85÷3.08≈0.3 for band B; the second weighted coefficient is the same, 0.5 for band C, 0.3 for band A, and 0.2 for band B.

[0137] Step 163: Based on the first weighting coefficient, enhance or attenuate the initial channel feature to obtain the target channel feature.

[0138] Among them, the first weighting coefficient is the coefficient determined in step 162 and used to adjust the initial channel characteristics; the enhancement or attenuation processing is to amplify or reduce the initial channel characteristics according to the coefficient; the target channel characteristics are channel characteristics that can better reflect their importance after processing; this step generates the target channel characteristics.

[0139] In the embodiment of the present application, the processing rule is first clarified as the initial channel feature multiplied by the first weighting coefficient; then each frequency band is processed, for example, the initial channel feature 5 of frequency band A is multiplied by 0.3, and 5×0.3=1.5 is obtained; 4 of frequency band B is multiplied by 0.3, and 4×0.3=1.2 is obtained; 6 of frequency band C is multiplied by 0.4, and 6×0.4=2.4 is obtained. These results are the target channel features.

[0140] Step 164: Based on the second weighting coefficient, perform regional weight allocation on the initial spatial features to obtain target spatial features.

[0141] Among them, the second weighting coefficient is the coefficient determined in step 162 and used to adjust the initial spatial features; the regional weight allocation is to allocate importance to the initial spatial features according to the coefficient; the target spatial features are the features that can better reflect the spatial importance after allocation; this step generates the target spatial features.

[0142] In the embodiment of the present application, the processing rule is first clarified as the initial spatial feature multiplied by the second weighting coefficient; then each frequency band is processed, for example, the initial spatial feature 3 of frequency band A is multiplied by 0.3, and 3×0.3=0.9 is obtained; 2 of frequency band B is multiplied by 0.2, and 2×0.2=0.4 is obtained; 5 of frequency band C is multiplied by 0.5, and 5×0.5=2.5 is obtained. These results are the target spatial features.

[0143] Step 165: Overlay and fuse the target channel features with the target spatial features to generate a weighted aggregate feature map group.

[0144] Among them, the target channel feature is the adjusted channel feature obtained in step 163; the target spatial feature is the adjusted spatial feature obtained in step 164; superposition fusion is the addition and integration of the two features; the weighted aggregated feature graph group is a set of graphs formed after fusion that reflects the importance of the features; this step generates a weighted aggregated feature graph group.

[0145] In an embodiment of the present application, the target channel features and target spatial features of each frequency band are first superimposed accordingly; for example, 1.5 plus 0.9 in frequency band A is obtained to obtain 1.5+0.9=2.4; 1.2 plus 0.4 in frequency band B is obtained to obtain 1.2+0.4=1.6; 2.4 plus 2.5 in frequency band C is obtained to obtain 2.4+2.5=4.9; and then these results are organized into charts to form a weighted aggregated feature graph group.

[0146] Step 166: Utilize the weighted aggregated feature map group to enhance the effective information of the high entropy value frequency band and suppress the noise interference of the low entropy value frequency band, so as to complete the pulse signal adaptive filtering based on the entropy value of the multi-band signal.

[0147] Among them, the weighted aggregated feature graph group is the integrated feature graph obtained in step 165; the high entropy frequency band is a frequency band with a large entropy value and a lot of effective information; the low entropy frequency band is a frequency band with a small entropy value and a lot of noise; the pulse signal adaptive filtering is a process of automatic adjustment to retain effective information and reduce noise; this step completes the adaptive filtering.

[0148] In an embodiment of the present application, high-entropy and low-entropy frequency bands are first distinguished from the weighted aggregated feature graph group, for example, the C band 4.9 corresponds to the high entropy value, and the B band 1.6 corresponds to the low entropy value; then, the effective information of the high-entropy frequency band is enhanced according to the graph group information, such as amplifying the C band signal; at the same time, the noise of the low-entropy frequency band is suppressed, such as reducing the B band signal, thereby completing adaptive filtering.

[0149] For example, when a system processes A, B, and C three-band signals, the initial channel features 5 and initial spatial features 3 of the A-band, the initial channel features 4 and initial spatial features 2 of the B-band, and the initial channel features 6 and initial spatial features 5 of the C-band are extracted from the aggregated feature map group. According to the multi-band signal entropy values of each frequency band (A: 1.03, B: 0.85, C: 1.20), the total entropy value 3.08 is calculated by combining the attention mechanism, the first weighting coefficient (A: 1.03 ÷ 3.08 ≈ 0.3, B: 0.85 ÷ 3.08 ≈ 0.3, C: 1.20 ÷ 3.08 ≈ 0.4) and the second weighting coefficient (A: 0.3, B: 0.2, C: 0.5) are determined. The initial channel features are processed according to the first weighting coefficient, A-band 5 × 0.3 = 1.5, B-band 4 × 0.3 = 1.2, and C-band 6 × 0.4 = 2.4, to obtain the target channel features. The initial spatial features are processed according to the second weighting coefficient, A-band 3 × 0.3 = 0.9, B-band 2 × 0.2 = 0.4, and C-band 5 × 0.5 = 2.5, to obtain the target spatial features. The target channel features and the target spatial features are superimposed, A-band 1.5 + 0.9 = 2.4, B-band 1.2 + 0.4 = 1.6, and C-band 2.4 + 2.5 = 4.9, to generate the weighted aggregated feature map group. Finally, the effective information of the C-band (high entropy value) is enhanced and the noise interference of the B-band (low entropy value) is suppressed through the map group, to complete the adaptive filtering of the pulse signal.

[0150] By performing steps 161-166, the embodiment of the application extracts the basic features from the aggregated feature map group, determines the feature weight by combining the signal complexity and the attention mechanism, highlights the important features by weighting adjustment, and then fuses the channel and spatial features to form a comprehensive chart. Finally, the effective information of the high-entropy value frequency band is enhanced and the noise of the low-entropy value frequency band is suppressed. The steps are closely connected, so that the filtering process can accurately adapt to the signal features of different frequency bands, effectively improving the clarity of the effective information in the signal and reducing irrelevant interference, so that the filtering result is more in line with the actual needs.

[0151] Figure 2 A structural diagram of a pulse signal adaptive filtering system based on multi-band signal entropy values provided by the embodiment of the application is shown in FIG. 1. Figure 2 As shown in FIG. 1, the system comprises: A receiving module 21 is configured to receive satellite communication signals of multiple different frequency bands, and extract the phase offset, amplitude value, and frequency value in the satellite communication signals.

[0152] A first generating module 22 is configured to generate a multi-dimensional feature group corresponding to each frequency band based on the phase offset, amplitude value, and frequency value, and extract the signal-to-noise ratio and frequency change of the corresponding frequency band from each multi-dimensional feature group.

[0153] The adjusting module 23 dynamically adjusts the filter characteristic parameters through a preset three-level tuning structure based on the signal-to-noise ratios and frequency variation amounts of all frequency bands, filters the satellite communication signals of different frequency bands based on the dynamically adjusted filter characteristic parameters, and obtains a filtered signal.

[0154] The transmission module 24 transmits the filtered signal to a federated learning framework, and calculates a multi-frequency signal entropy value based on the federated learning framework and the multi-dimensional feature group.

[0155] The second generation module 25 optimizes adaptive filter parameters based on the multi-frequency signal entropy value, globally aggregates the optimized adaptive filter parameters through a differential privacy protection rule, and generates an aggregated feature map group.

[0156] The weighting module 26 weights the aggregated feature map group based on a channel and spatial attention mechanism to enhance effective information of a high-entropy value frequency band and suppress noise interference of a low-entropy value frequency band, and completes the multi-frequency signal entropy value-based adaptive filtering of the pulse signal.

[0157] Figure 2 The multi-frequency signal entropy value-based adaptive filtering system of the pulse signal can perform Figure 1 The implementation principle and technical effects of the multi-frequency signal entropy value-based adaptive filtering method of the pulse signal of the embodiments will not be described again. For the specific manner in which each module, unit of the multi-frequency signal entropy value-based adaptive filtering system of the pulse signal in the above embodiments performs operations, the detailed description has been made in the embodiments related to the method, and will not be described in detail here.

[0158] In one possible design, Figure 2 The multi-frequency signal entropy value-based adaptive filtering system of the pulse signal of the embodiments can be implemented as a computing device, such as a server. Figure 3 As shown, the computing device can include a storage component 31 and a processing component 32.

[0159] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32.

[0160] The processing component 32 is configured to execute the following process: receiving satellite communication signals of multiple different frequency bands, extracting phase offsets, amplitude values, and frequency values ​​from the satellite communication signals; generating a multidimensional feature group corresponding to each frequency band based on the phase offsets, amplitude values, and frequency values, and extracting the signal-to-noise ratio and frequency variation of the corresponding frequency band from each multidimensional feature group; dynamically adjusting the filtering characteristic parameters based on the signal-to-noise ratio and frequency variation of all frequency bands through a preset three-level tuning structure, filtering the satellite communication signals of different frequency bands based on the dynamically adjusted filtering characteristic parameters to obtain filtered signals; transmitting the filtered signals to a federated learning framework, calculating the multi-band signal entropy values ​​based on the federated learning framework and the multi-dimensional feature group; optimizing the adaptive filtering parameters based on the multi-band signal entropy values, and globally aggregating the optimized adaptive filtering parameters through differential privacy protection rules to generate an aggregated feature map group; and weighting the aggregated feature map group based on a channel and spatial attention mechanism to enhance the effective information of the high-entropy frequency band and suppress the noise interference of the low-entropy frequency band, thereby completing the pulse signal adaptive filtering based on the multi-band signal entropy values.

[0161] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above method.

[0162] The storage component 31 is configured to store various types of data to support operations on the terminal. The storage component can be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as random access memory (RAM), static random access memory (SRAM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0163] Of course, a computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0164] The input / output interface provides an interface between the processing component and the peripheral interface module, which can be an output device, an input device, etc.

[0165] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0166] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0167] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The embodiment shown is a pulse signal adaptive filtering method based on multi-band signal entropy values.

[0168] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0169] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0170] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0171] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A pulse signal adaptive filtering method based on multi-band signal entropy value, characterized in that: include: receiving a plurality of satellite communication signals in different frequency bands, and extracting phase offset, amplitude and frequency values ​​from the satellite communication signals; generating a multidimensional feature group corresponding to each frequency band based on the phase offset, the amplitude value, and the frequency value, and extracting a signal-to-noise ratio and a frequency variation of the corresponding frequency band from each multidimensional feature group; Based on the signal-to-noise ratio and the frequency variation of all frequency bands, dynamically adjusting the filter characteristic parameters through a preset three-stage tuning structure, and filtering the satellite communication signals of the different frequency bands based on the dynamically adjusted filter characteristic parameters to obtain filtered signals; Transmitting the filtered signal to a federated learning framework, and calculating a multi-band signal entropy value based on the federated learning framework and the multidimensional feature set; Based on the multi-band signal entropy values, optimizing the adaptive filtering parameters, and globally aggregating the optimized adaptive filtering parameters through differential privacy protection rules to generate an aggregated feature map group; Based on the channel and spatial attention mechanism, the aggregated feature map group is weighted to enhance the effective information of the high entropy frequency band and suppress the noise interference of the low entropy frequency band, thereby completing the adaptive filtering of the pulse signal based on the entropy value of the multi-band signal.

2. The method according to claim 1, characterized in that The method dynamically adjusts the filter characteristic parameters based on the signal-to-noise ratio and the frequency variation of all frequency bands through a preset three-level tuning structure, including: Based on the signal-to-noise ratios of all frequency bands, filter bandwidth adjustment parameters of corresponding frequency bands are calculated according to a preset inverse proportional relationship to construct a first-level tuning structure; Based on the frequency changes of all frequency bands, center frequency adjustment parameters of corresponding frequency bands are calculated according to a preset proportional correspondence to construct a second-level tuning structure; Integrating the signal-to-noise ratios and the frequency variations of all frequency bands, and obtaining a signal comprehensive evaluation parameter based on the integration result through a preset weighted calculation method; Based on the comprehensive signal evaluation parameters, the signal gain adjustment parameters of the corresponding frequency band are determined according to a preset corresponding relationship to construct a third-level tuning structure; Integrating the first-stage tuning structure, the second-stage tuning structure, and the third-stage tuning structure to obtain a preset three-stage tuning structure; Based on the three-stage tuning structure, the filter bandwidth adjustment parameter, the center frequency adjustment parameter and the signal gain adjustment parameter are mapped to the filter characteristics of the corresponding frequency band to dynamically adjust the filter characteristic parameters to obtain the dynamically adjusted filter characteristic parameters.

3. The method according to claim 2, characterized in that The method of mapping the filter bandwidth adjustment parameter, the center frequency adjustment parameter, and the signal gain adjustment parameter to the filter characteristics of the corresponding frequency band based on the three-stage tuning structure to dynamically adjust the filter characteristic parameters to obtain the dynamically adjusted filter characteristic parameters includes: Determine a basic parameter set for each frequency band, wherein the basic parameter set includes a basic bandwidth parameter, a basic center frequency parameter, and a basic gain parameter; Based on a preset bandwidth mapping rule, the filter bandwidth adjustment parameter in the three-stage tuning structure is combined with the basic bandwidth parameter of the corresponding frequency band to obtain a dynamic bandwidth parameter; Based on a preset frequency mapping rule, the center frequency adjustment parameter in the three-level tuning structure is combined with the basic center frequency parameter of the corresponding frequency band to obtain a dynamic center frequency parameter; Based on a preset gain mapping rule, the signal gain adjustment parameter in the three-stage tuning structure is combined with the basic gain parameter of the corresponding frequency band to obtain a dynamic gain parameter; The dynamic bandwidth parameter, the dynamic center frequency parameter, and the dynamic gain parameter are mapped to the filter characteristic of the corresponding frequency band to generate dynamically adjusted filter characteristic parameters of the corresponding frequency band.

4. The method according to claim 1, wherein The step of transmitting the filtered signal to a federated learning framework and calculating a multi-band signal entropy value based on the federated learning framework and the multi-dimensional feature group includes: Based on a preset frequency band division standard, the filtered signal is divided into a plurality of sub-signals, and the plurality of sub-signals are respectively transmitted to nodes corresponding to each frequency band in the federated learning framework, so that after each node receives the sub-signal of the corresponding frequency band, the sub-signal is associated with the multidimensional feature group, and based on the association result, characteristic parameters are calculated on the associated data of the sub-signal and the multidimensional feature group according to a preset local calculation rule to obtain signal characteristic distribution parameters of the corresponding frequency band; Based on the federated learning framework and the preset cross-node collaboration rules, the signal feature distribution parameters of the corresponding frequency bands of each node are integrated to form a global feature distribution parameter; The global characteristic distribution parameters are calculated according to a preset entropy value calculation rule to obtain the multi-band signal entropy value.

5. The method according to claim 4, characterized in that The method of calculating characteristic parameters of the sub-signals and the associated data in the multi-dimensional feature group based on the correlation result and according to a preset local calculation rule to obtain the signal characteristic distribution parameters of the corresponding frequency band includes: Comparing the phase offset of the associated data in the multidimensional feature group with a preset reference phase value to generate a phase difference value sequence; Extracting a maximum difference and a minimum difference from the phase difference sequence, and calculating a phase fluctuation range parameter based on the maximum difference and the minimum difference; Based on the time domain distribution statistical rule in the preset local calculation rule, the amplitude value of the correlation data is calculated to obtain the amplitude change rate parameter; Calculating the difference between the frequency value in the associated data and a preset reference frequency value to obtain a frequency offset parameter; Constructing a characteristic parameter set according to the phase fluctuation range parameter, the amplitude change rate parameter, and the frequency offset parameter; A distribution characteristic analysis is performed on the characteristic parameter set, and based on the result of the distribution characteristic analysis, a signal characteristic distribution parameter of the corresponding frequency band is generated.

6. The method according to claim 1, characterized in that The method of optimizing the adaptive filtering parameters based on the multi-band signal entropy values, and globally aggregating the optimized adaptive filtering parameters through differential privacy protection rules to generate an aggregated feature map group includes: Determining the adjustment weight of the adaptive filtering parameters according to the entropy value of the multi-band signal; Based on the adjustment weights, the initial adaptive filtering parameters are optimized step by step to obtain optimized adaptive filtering parameters; Based on the differential privacy protection rule, the privacy perturbation value that meets the preset threshold condition is superimposed on the optimized adaptive filtering parameters to obtain the target adaptive filtering parameters; Based on a preset parameter fusion ratio, the target adaptive filtering parameters are input into an aggregation node in the federated learning framework and then globally integrated to generate global parameters; The global parameters are reorganized according to preset feature dimensions to generate an aggregated feature map group.

7. The method according to claim 1, characterized in that The channel- and space-based attention mechanism weights the aggregated feature map group to enhance the effective information of the high-entropy frequency band and suppress the noise interference of the low-entropy frequency band, completing the pulse signal adaptive filtering based on the entropy value of the multi-band signal, including: Extracting initial channel features and initial spatial features corresponding to each frequency band from the aggregated feature map group; Determining a first weighting coefficient and a second weighting coefficient for the initial channel feature and the initial spatial feature, respectively, based on the multi-band signal entropy value of each frequency band and an attention mechanism; Based on the first weighting coefficient, the initial channel feature is enhanced or attenuated to obtain a target channel feature; Based on the second weighting coefficient, performing regional weight allocation on the initial spatial feature to obtain a target spatial feature; Superimposing and fusing the target channel features with the target spatial features to generate a weighted aggregate feature map group; The weighted aggregated feature map group is used to enhance the effective information of the high entropy value frequency band and suppress the noise interference of the low entropy value frequency band, so as to complete the pulse signal adaptive filtering based on the entropy value of the multi-band signal.

8. A pulse signal adaptive filtering system based on multi-band signal entropy value, characterized in that: include: A receiving module is used to receive satellite communication signals of multiple different frequency bands and extract phase offset, amplitude and frequency values ​​from the satellite communication signals; A first generating module is configured to generate a multidimensional feature group corresponding to each frequency band based on the phase offset, the amplitude value, and the frequency value, and extract the signal-to-noise ratio and frequency variation of the corresponding frequency band from each multidimensional feature group; an adjustment module, configured to dynamically adjust filtering characteristic parameters based on the signal-to-noise ratio and the frequency variation of all frequency bands through a preset three-stage tuning structure, and filter the satellite communication signals of the different frequency bands based on the dynamically adjusted filtering characteristic parameters to obtain filtered signals; a transmission module, configured to transmit the filtered signal to a federated learning framework, and calculate a multi-band signal entropy value based on the federated learning framework and the multidimensional feature set; A second generation module is configured to optimize adaptive filtering parameters based on the multi-band signal entropy values, and globally aggregate the optimized adaptive filtering parameters using differential privacy protection rules to generate an aggregated feature map group; The weighting module is used to weight the aggregated feature map group based on the channel and space attention mechanism to enhance the effective information of the high entropy frequency band and suppress the noise interference of the low entropy frequency band, thereby completing the pulse signal adaptive filtering based on the entropy value of the multi-band signal.

9. A computing device, characterized in that It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a pulse signal adaptive filtering method based on multi-band signal entropy value as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, the method for adaptive filtering of pulse signals based on multi-band signal entropy values ​​according to any one of claims 1 to 7 is implemented.

Images