Strike accuracy analysis method and device based on electromagnetic emission
Through the acquisition of electromagnetic emission factors and data noise reduction processing, combined with electromagnetic signal characteristics and accuracy indicators, an accuracy analysis model is constructed to determine the optimal electromagnetic emission parameters and strike strategies, which solves the problem of low electromagnetic emission strike accuracy and realizes the complex environment Efficient and precise strikes.
Patent Information
- Application Number
- CN202410450426.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-15
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-04-15
AI Technical Summary
The strike accuracy based on electromagnetic emission is low, and due to the complexity of the electromagnetic environment and noise interference, it is difficult to achieve accurate target locking and tracking in long distances or complex environments.
By determining the electromagnetic emission factors, collecting strike data and performing noise reduction processing, extracting electromagnetic signal characteristics and accuracy indicators, building an accuracy analysis model, determining the optimal electromagnetic emission parameters and strike strategies, and improving strike accuracy.
It effectively improves the strike accuracy of electromagnetic emission, enhances the target locking and tracking capabilities in complex environments, and improves the efficiency and accuracy in areas such as energy mining and resource exploration.
Smart Images

Figure QLYQS_1 
Figure QLYQS_10 
Figure QLYQS_21
Abstract
Description
Technical Field
[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a strike accuracy analysis method and device based on electromagnetic emission. Background Art
[0002] In the engineering field, with the continuous advancement and innovation of science and technology, it is becoming increasingly important to improve the accuracy of strikes based on electromagnetic launch. By utilizing advanced electromagnetic technology, more accurate target locking and strikes can be achieved, which is of great significance to fields such as energy mining, resource exploration and environmental monitoring. For example, in energy mining, precise strikes can help improve mining efficiency and reduce the waste of energy resources; in resource exploration, accurate target locking can promote the accuracy and efficiency of exploration work.
[0003] Due to the complex and changeable electromagnetic environment, there are many interference signals and noise, which affect the precise positioning and tracking of the target. In addition, the propagation characteristics of electromagnetic waves limit the strike accuracy at long distances or in complex environments. Therefore, how to improve the strike accuracy based on electromagnetic emission has become an urgent problem to be solved. Summary of the invention
[0004] The present invention provides a method and device for analyzing strike accuracy based on electromagnetic emission, the main purpose of which is to solve the problem of low strike accuracy based on electromagnetic emission.
[0005] To achieve the above-mentioned purpose, the present invention provides a method for analyzing strike accuracy based on electromagnetic emission, comprising:
[0006] determining electromagnetic emission factors of a strike target, and collecting strike data of the strike target according to the electromagnetic emission factors;
[0007] Performing data denoising on the strike data to obtain denoised data of the strike data;
[0008] Extracting electromagnetic signal characteristics of the strike target according to the noise reduction data, and generating an accuracy index of the strike target according to the noise reduction data;
[0009] A training set of a pre-constructed precision analysis model is generated according to the electromagnetic signal characteristics and the precision index, and a loss function of the pre-constructed precision analysis model is determined, wherein the loss function is:
[0010]
[0011] Wherein, J(Θ) is the loss function value of the pre-built accuracy analysis model, Θ is the model parameter of the pre-built accuracy analysis model, m is the number of training samples in the training set, i is the sample identifier of the training sample in the training set, and y (i)is the sample label of the i-th sample, is the model output representing the i-th sample;
[0012] Performing model training on the pre-built accuracy analysis model using the training set and the loss function to obtain a trained accuracy analysis model;
[0013] The trained precision analysis model is used to determine the optimal electromagnetic emission parameters for the strike target, and a strike strategy for the target to be struck is generated based on the optimal electromagnetic emission parameters.
[0014] Optionally, the electromagnetic emission factors are: electromagnetic signal frequency, electromagnetic signal amplitude and electromagnetic signal phase.
[0015] Optionally, collecting the strike data of the strike target according to the electromagnetic emission factor includes:
[0016] Determining the strike parameters of the strike target according to the electromagnetic emission factors;
[0017] Performing a target strike on the strike target according to the strike parameter to obtain reflection data of the strike target;
[0018] The striking data of the striking target is generated according to the striking parameters and the launching data.
[0019] Optionally, performing data denoising on the strike data to obtain denoised data of the strike data includes:
[0020] Performing wavelet transform on the attack data to obtain wavelet coefficients of the attack data;
[0021] Performing soft threshold processing on the wavelet coefficients according to a preset coefficient threshold to obtain effective coefficients of the wavelet coefficients;
[0022] Perform an inverse wavelet transform on the effective coefficients to obtain transformation data of the effective coefficients, and determine that the transformation data is noise reduction data of the attack data.
[0023] Optionally, performing data denoising on the strike data to obtain denoised data of the strike data includes:
[0024] The adjustment coefficient of the attack data is generated using a preset adjustment coefficient generation algorithm, wherein the preset adjustment coefficient generation algorithm is:
[0025] w(n+1)=w(n)+μe(n)X(n)
[0026] Wherein, w(n+1) is the adjustment coefficient vector at the n+1th moment, w(n) is the adjustment coefficient vector at the nth moment, μ is the learning rate, e(n) is the error between the estimated output and the actual output corresponding to the attack data at the nth moment, X(n) is the attack data at the nth moment, and n is the time corresponding to the attack data;
[0027] The striking data is corrected using the adjustment coefficient to obtain corrected data of the striking data, and the corrected data is determined to be noise reduction data of the striking data.
[0028] Optionally, extracting the electromagnetic signal characteristics of the strike target according to the noise reduction data includes:
[0029] Extracting frequency characteristics of the strike target according to the noise reduction data;
[0030] generating an amplitude feature of the strike target according to the noise reduction data;
[0031] Generating a phase feature of the strike target according to the noise reduction data;
[0032] The electromagnetic signal characteristics of the strike target are generated according to the frequency characteristics, the amplitude characteristics and the phase characteristics.
[0033] Optionally, generating the accuracy index of the strike target according to the noise reduction data includes:
[0034] Performing data screening on the noise reduction data to obtain target data of the noise reduction data;
[0035] Generating a strike evaluation value of the strike target according to the target data, wherein the strike evaluation value includes: hit rate, accuracy and strike deviation;
[0036] An accuracy index of the strike target is generated according to the strike evaluation value.
[0037] Optionally, generating a training set of a pre-built accuracy analysis model according to the electromagnetic signal characteristics and the accuracy index includes:
[0038] Establishing a correlation relationship between the electromagnetic signal characteristics and the accuracy index;
[0039] generating a precision label for the precision indicator;
[0040] A training set of a pre-built precision analysis model is generated according to the electromagnetic signal characteristics, the association relationship and the precision label.
[0041] Optionally, the determining the optimal electromagnetic emission parameters for striking the target by using the trained accuracy analysis model includes:
[0042] Acquiring input features of the strike target;
[0043] Generate the strike accuracy of the strike target according to the input feature and the trained accuracy analysis model;
[0044] The optimal electromagnetic emission parameters for striking the target are generated according to the striking accuracy.
[0045] In order to solve the above problems, the present invention also provides a strike accuracy analysis device based on electromagnetic emission, the device comprising:
[0046] A data acquisition module, used to determine the electromagnetic emission factors of the strike target, and to collect strike data of the strike target according to the electromagnetic emission factors;
[0047] A data denoising module, used for performing data denoising on the attack data to obtain denoised data of the attack data;
[0048] An accuracy index generating module, used for extracting the electromagnetic signal characteristics of the strike target according to the noise reduction data, and generating the accuracy index of the strike target according to the noise reduction data;
[0049] A training set generation module is used to generate a training set of a pre-built accuracy analysis model according to the electromagnetic signal characteristics and the accuracy index, and determine the loss function of the pre-built accuracy analysis model, wherein the loss function is:
[0050]
[0051] Wherein, J(Θ) is the loss function value of the pre-built accuracy analysis model, Θ is the model parameter of the pre-built accuracy analysis model, m is the number of training samples in the training set, i is the sample identifier of the training sample in the training set, and y (i) is the sample label of the i-th sample, is the model output representing the i-th sample;
[0052] A model training module, used to perform model training on the pre-built accuracy analysis model using the training set and the loss function to obtain a trained accuracy analysis model;
[0053] The strike strategy generation module is used to determine the optimal electromagnetic emission parameters of the strike target using the trained precision analysis model, and generate a strike strategy for the target to be struck based on the optimal electromagnetic emission parameters.
[0054] The present invention achieves accurate understanding of the target by determining the electromagnetic emission factors of the target and obtaining clear data through noise reduction processing. Secondly, the precision analysis model is trained using the extracted electromagnetic signal characteristics and precision indicators, so that the target characteristics can be deeply analyzed and the optimal strike parameters can be learned. Finally, the optimal electromagnetic emission parameters are determined and the optimal strike strategy is generated through the model, which effectively improves the strike accuracy. Therefore, the present invention proposes a strike accuracy analysis method and device based on electromagnetic emission, which can be used to improve the strike accuracy of electromagnetic emission. BRIEF DESCRIPTION OF THE DRAWINGS
[0055] Figure 1 A schematic diagram of a flow chart of a method for analyzing strike accuracy based on electromagnetic emission provided by an embodiment of the present invention;
[0056] Figure 2 A schematic diagram of a process for denoising strike data provided by an embodiment of the present invention;
[0057] Figure 3 A schematic diagram of a process for extracting electromagnetic signal characteristics of a strike target provided by an embodiment of the present invention;
[0058] Figure 4 A functional module diagram of a strike accuracy analysis device based on electromagnetic emission provided by an embodiment of the present invention;
[0059] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION
[0060] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.
[0061] The embodiment of the present application provides a strike precision analysis method based on electromagnetic emission. The execution subject of the strike precision analysis method based on electromagnetic emission includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided in the embodiment of the present application. In other words, the strike precision analysis method based on electromagnetic emission can be executed by software or hardware installed on a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.
[0062] Reference Figure 1FIG. 1 is a flow chart of a method for analyzing strike accuracy based on electromagnetic emission provided by an embodiment of the present invention. In this embodiment, the method for analyzing strike accuracy based on electromagnetic emission includes:
[0063] S1. Determine the electromagnetic emission factors of the strike target, and collect strike data of the strike target according to the electromagnetic emission factors.
[0064] In the embodiment of the present invention, the electromagnetic emission factors are: electromagnetic signal frequency, electromagnetic signal amplitude and electromagnetic signal phase.
[0065] In detail, the electromagnetic emission factors directly affect the characteristics and propagation behavior of electromagnetic signals; the strike target refers to the target object that needs to be struck or attacked, which may be a target, target object or other specific object; the strike data refers to the data obtained after striking or attacking the strike target through electromagnetic emission and the data at the time of striking, which may include reflection data, emission data and other information.
[0066] In detail, the frequency of the electromagnetic signal determines the periodicity of the signal waveform, and signals of different frequencies will be affected by different attenuation and scattering during the propagation process.
[0067] In detail, the electromagnetic signal amplitude represents the strength of the signal, which can reflect the reflection ability or radiation intensity of the target.
[0068] In detail, the phase of the electromagnetic signal describes the deviation of the signal waveform, and signals with different phases can be used to distinguish different targets or extract specific feature information.
[0069] In the embodiment of the present invention, the collecting the strike data of the strike target according to the electromagnetic emission factor includes:
[0070] Determining the strike parameters of the strike target according to the electromagnetic emission factors;
[0071] Performing a target strike on the strike target according to the strike parameter to obtain reflection data of the strike target;
[0072] The striking data of the striking target is generated according to the striking parameters and the launching data.
[0073] In detail, determining the strike parameters of the strike target based on the electromagnetic emission factors requires determining the relevant parameters of the electromagnetic emission in advance, such as frequency, amplitude and phase, and determining the specific parameter settings required for striking the target based on these parameters, such as the characteristics of the radio frequency waveform.
[0074] In detail, the target is struck by setting the strike parameters, the reflection of the electromagnetic wave by the target is observed, and the reflection data is recorded. The data may include information such as the reflectivity and scattering characteristics of the target object.
[0075] In detail, the previously collected strike parameters and reflection data are combined to generate complete strike data, which can be used for subsequent analysis and processing, such as extracting target features and evaluating strike effects.
[0076] S2. Performing data denoising on the attack data to obtain denoised data of the attack data.
[0077] In the embodiment of the present invention, Figure 2 As shown, the step of performing data denoising on the strike data to obtain denoised data of the strike data includes:
[0078] S21, performing wavelet transform on the attack data to obtain wavelet coefficients of the attack data;
[0079] S22, performing soft threshold processing on the wavelet coefficients according to a preset coefficient threshold to obtain effective coefficients of the wavelet coefficients;
[0080] S23, performing inverse wavelet transform on the effective coefficients to obtain transformed data of the effective coefficients, and determining that the transformed data is denoised data of the attack data.
[0081] In detail, performing wavelet transform on the strike data refers to decomposing the strike data into sub-bands of different scales and frequencies.
[0082] In detail, by analyzing the wavelet coefficients, appropriate thresholds are determined to determine which wavelet coefficients represent signals and which represent noise. A commonly used threshold processing method is soft threshold processing, that is, setting wavelet coefficients less than the threshold to zero and retaining coefficients greater than the threshold.
[0083] In detail, the inverse wavelet transform of the effective coefficients refers to reconstructing the processed wavelet coefficients into a denoised signal, that is, reconstructing the effective coefficients into a denoised signal.
[0084] Furthermore, assuming that there is a set of strike data containing noise, wavelet transform will be used to perform noise reduction on these data: first, Daubechies wavelet can be selected as the wavelet basis function, and the strike data can be decomposed in the time domain and frequency domain to obtain wavelet coefficients at various scales; then, by analyzing the wavelet coefficients, an appropriate threshold is determined to distinguish between signals and noise. A threshold can be selected to perform soft threshold processing on wavelet coefficients less than the threshold, that is, these coefficients are set to zero, and coefficients greater than the threshold are retained, for example: 3 times the standard deviation of the noise is used as the threshold; finally, after threshold processing, the processed wavelet coefficients are inversely transformed to reconstruct them into noise-reduced signals.
[0085] In detail, through the above steps, the noise in the strike data can be removed and the noise-reduced signal can be obtained, so that the processed data is clearer and more conducive to subsequent analysis and application.
[0086] In the embodiment of the present invention, performing data denoising on the strike data to obtain denoised data of the strike data includes:
[0087] The adjustment coefficient of the attack data is generated using a preset adjustment coefficient generation algorithm, wherein the preset adjustment coefficient generation algorithm is:
[0088] w(n+1)=w(n)+μe(n)X(n)
[0089] Wherein, w(n+1) is the adjustment coefficient vector at the n+1th moment, w(n) is the adjustment coefficient vector at the nth moment, μ is the learning rate, e(n) is the error between the estimated output and the actual output corresponding to the attack data at the nth moment, X(n) is the attack data at the nth moment, and n is the time corresponding to the attack data;
[0090] The striking data is corrected using the adjustment coefficient to obtain corrected data of the striking data, and the corrected data is determined to be noise reduction data of the striking data.
[0091] In detail, it is assumed that there is a set of strike data containing noise as an observation signal, wherein a preset adjustment coefficient generation algorithm can adaptively adjust the filter coefficient according to the characteristics of the observation signal to suppress the noise while retaining the main components of the signal.
[0092] In detail, the preset adjustment coefficient generation algorithm can generate different adjustment coefficient vectors according to the signal characteristics observed in real time, so as to better adapt to the changes in the signal and the characteristics of the noise, and have a better ability to suppress noise.
[0093] In detail, using a preset adjustment coefficient generation algorithm to generate an adjustment coefficient for strike data means first setting a learning rate and an initial adjustment coefficient vector according to the preset algorithm, then processing the strike data, calculating the error between the estimated output and the actual output, and inputting it together with the strike data into the adjustment coefficient generation algorithm to obtain a new adjustment coefficient vector. This process can be repeated multiple times until the optimal adjustment coefficient vector is obtained.
[0094] Furthermore, the obtained optimal adjustment coefficient vector is applied to the strike data to perform data correction, remove noise and interference therein, and obtain corrected data, namely, noise-reduced data.
[0095] S3. Extracting electromagnetic signal characteristics of the strike target according to the noise reduction data, and generating an accuracy index of the strike target according to the noise reduction data.
[0096] In the embodiment of the present invention, Figure 3 As shown, the step of extracting the electromagnetic signal characteristics of the strike target according to the noise reduction data includes:
[0097] S31, extracting the frequency characteristics of the strike target according to the noise reduction data;
[0098] S32, generating an amplitude feature of the strike target according to the noise reduction data;
[0099] S33, generating a phase feature of the strike target according to the noise reduction data;
[0100] S34, generating the electromagnetic signal characteristics of the strike target according to the frequency characteristics, the amplitude characteristics and the phase characteristics.
[0101] In detail, the frequency characteristics of the strike target extracted according to the noise reduction data can be extracted by performing spectrum analysis or other frequency domain processing methods on the noise-reduced data, including main frequency components, spectrum shape and other information.
[0102] In detail, generating the amplitude characteristics of the strike target according to the noise reduction data means that the amplitude characteristics of the strike target can be calculated according to the noise-reduced data, which can reflect the energy distribution and amplitude change of the signal.
[0103] In detail, generating the phase characteristics of the strike target based on the noise reduction data means using phase demodulation or other phase extraction methods to obtain the phase characteristics of the strike target from the noise reduction data, which helps to capture the phase change information of the signal.
[0104] In detail, generating the electromagnetic signal characteristics of the strike target based on the frequency characteristics, the amplitude characteristics and the phase characteristics means combining the extracted frequency characteristics, amplitude characteristics and phase characteristics to form a complete electromagnetic signal characteristic, and these characteristic combinations can more comprehensively describe the electromagnetic signal characteristics of the strike target.
[0105] In the embodiment of the present invention, the step of generating the accuracy index of the strike target according to the noise reduction data includes:
[0106] Performing data screening on the noise reduction data to obtain target data of the noise reduction data;
[0107] Generating a strike evaluation value of the strike target according to the target data, wherein the strike evaluation value includes: hit rate, accuracy and strike deviation;
[0108] An accuracy index of the strike target is generated according to the strike evaluation value.
[0109] In detail, the noise reduction data refers to the data obtained by processing the collected strike data and removing the noise and interference therein; the target data refers to the noise reduction data retained after screening and with clear target information; the strike evaluation value is used to evaluate the indicators of the strike target, including hit rate, accuracy and strike deviation.
[0110] In detail, the strike evaluation value of the strike target is generated according to the target data in order to evaluate the strike target using the target data. According to the actual situation, the hit rate (i.e. the ratio of the number of hits to the total number of hits), precision (i.e. the accuracy of the strike target) and strike deviation (i.e. the distance between the actual strike position and the target position, etc.) and other indicators can be calculated to evaluate the effect and accuracy of the strike.
[0111] Furthermore, generating the accuracy index of the strike target based on the strike evaluation value means integrating the calculated strike evaluation values to generate the accuracy index of the strike target. Indicators such as hit rate, accuracy and strike deviation can be comprehensively considered, and corresponding evaluation methods and indicator systems can be formulated according to actual needs to evaluate the accuracy level of the strike target.
[0112] S4. Generate a training set of a pre-constructed accuracy analysis model according to the electromagnetic signal characteristics and the accuracy index, and determine the loss function of the pre-constructed accuracy analysis model.
[0113] In an embodiment of the present invention, generating a training set of a pre-built accuracy analysis model according to the electromagnetic signal characteristics and the accuracy index includes:
[0114] Establishing a correlation relationship between the electromagnetic signal characteristics and the accuracy index;
[0115] generating a precision label for the precision indicator;
[0116] A training set of a pre-built precision analysis model is generated according to the electromagnetic signal characteristics, the association relationship and the precision label.
[0117] In detail, the electromagnetic signal characteristics refer to signal characteristics related to the strike target collected by the electromagnetic sensor, which may include frequency, amplitude, phase and other information.
[0118] In detail, the pre-built accuracy analysis model refers to a model established based on electromagnetic signal characteristics and accuracy indicators for predicting the accuracy of hitting a target.
[0119] In detail, by analyzing the existing electromagnetic signal characteristic data and the corresponding accuracy indicators and establishing the correlation relationship between them, a machine learning algorithm or a statistical method can be used to perform correlation analysis.
[0120] In detail, generating the precision label of the precision index means converting the precision index into discrete labels, such as excellent, qualified, unqualified, etc., as labels for training samples, based on existing precision index data and given a threshold or range.
[0121] In detail, generating a training set for a pre-built precision analysis model based on electromagnetic signal characteristics, correlation relationships and precision labels means forming training samples consisting of labeled electromagnetic signal characteristics and corresponding precision labels as the training set for the pre-built precision analysis model. These training samples will be used to train the model so that it can predict the accuracy of striking targets based on electromagnetic signal characteristics.
[0122] In detail, the loss function is:
[0123]
[0124] Wherein, J(Θ) is the loss function value of the pre-built accuracy analysis model, Θ is the model parameter of the pre-built accuracy analysis model, m is the number of training samples in the training set, i is the sample identifier of the training sample in the training set, and y (i) is the sample label of the i-th sample, is the model output representing the i-th sample.
[0125] In detail, the loss function can be interpreted as: for each training sample i, calculate its predicted value With the actual label y (i)The difference between them is taken into account in the calculation of the loss function. When the actual label is 1, the first term of the loss function measures the deviation between the model’s prediction of sample i and the actual value of 1. When the actual label is 0, the second term of the loss function measures the deviation between the model’s prediction of sample i and the actual value of 0. The entire loss function calculates the average value of this prediction deviation over all samples.
[0126] Finally, by minimizing this loss function, the optimal model parameter Θ can be found so that the model's prediction results are as close to the actual labels as possible, thereby improving the accuracy of the model.
[0127] S5. Perform model training on the pre-built precision analysis model using the training set and the loss function to obtain a trained precision analysis model.
[0128] In an embodiment of the present invention, the use of the training set and the loss function to perform model training on the pre-built precision analysis model to obtain a trained precision analysis model refers to inputting the training set into the pre-built precision analysis model, calculating the loss function according to the model prediction results and the actual labels, updating the model parameters using the back propagation algorithm, and continuously iterating and optimizing the model until the loss function converges to a minimum value. After multiple iterative trainings, a trained precision analysis model is obtained, which has the ability to predict the accuracy of striking targets based on electromagnetic signal characteristics.
[0129] In detail, the training set refers to a dataset used to train a pre-built accuracy analysis model, which contains annotated electromagnetic signal features and corresponding accuracy labels.
[0130] In detail, the loss function is a function used to evaluate the error between the model's prediction results and the actual results. During the model training process, the model parameters are optimized by minimizing the loss function so that the model can more accurately predict the accuracy of hitting the target.
[0131] S6. Determine the optimal electromagnetic emission parameters for the target by using the trained precision analysis model, and generate a strike strategy for the target to be struck based on the optimal electromagnetic emission parameters.
[0132] In an embodiment of the present invention, the step of determining the optimal electromagnetic emission parameters for striking the target by using the trained accuracy analysis model includes:
[0133] Acquiring input features of the strike target;
[0134] Generate the strike accuracy of the strike target according to the input feature and the trained accuracy analysis model;
[0135] The optimal electromagnetic emission parameters for striking the target are generated according to the striking accuracy.
[0136] In detail, the input feature refers to an electromagnetic emission parameter, and at the same time provides a parameter range of the electromagnetic emission parameter of a currently available electromagnetic emission device.
[0137] In detail, the acquired input features (i.e., electromagnetic emission parameters) are input into the trained accuracy analysis model, and the accuracy level of the strike target is evaluated based on the strike accuracy evaluation value predicted by the model.
[0138] Suppose you want to strike a high-speed drone, you first need to obtain information such as the drone's position and speed, as well as the electromagnetic emission parameter range of the available electromagnetic emission equipment. Then, using the trained accuracy analysis model, input this information to predict the strike accuracy evaluation value of the drone. Then, by analyzing the strike accuracy evaluation value, generate the optimal electromagnetic emission parameter combination to achieve the best strike effect.
[0139] In an embodiment of the present invention, the generation of a strike strategy for a target to be struck based on the optimal electromagnetic emission parameters means that after determining the optimal electromagnetic emission parameters, a specific strike method needs to be determined so as to achieve an effective strike on the target. For example, by specifying parameters such as the frequency, power and phase of the electromagnetic emission, the electromagnetic waves can interfere with or destroy the target at the right time and in an appropriate manner.
[0140] The present invention achieves accurate understanding of the target by determining the electromagnetic emission factors of the target and obtaining clear data through noise reduction processing. Secondly, the precision analysis model is trained using the extracted electromagnetic signal characteristics and precision indicators, so that the target characteristics can be deeply analyzed and the optimal strike parameters can be learned. Finally, the optimal electromagnetic emission parameters are determined and the optimal strike strategy is generated through the model, which effectively improves the strike accuracy. Therefore, the present invention proposes a strike accuracy analysis method based on electromagnetic emission, which can be used to improve the strike accuracy of electromagnetic emission.
[0141] like Figure 4 1 is a functional module diagram of a strike accuracy analysis device based on electromagnetic emission provided by one embodiment of the present invention.
[0142] The strike precision analysis device 100 based on electromagnetic emission of the present invention can be installed in an electronic device. According to the functions to be implemented, the strike precision analysis device 100 based on electromagnetic emission can include a data acquisition module 101, a data noise reduction module 102, an accuracy index generation module 103, a training set generation module 104, a model training module 105 and a strike strategy generation module 106. The module of the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.
[0143] In this embodiment, the functions of each module / unit are as follows:
[0144] The data acquisition module 101 is used to determine the electromagnetic emission factors of the strike target and collect strike data of the strike target according to the electromagnetic emission factors;
[0145] The data denoising module 102 is used to perform data denoising on the attack data to obtain denoised data of the attack data;
[0146] The accuracy index generating module 103 is used to extract the electromagnetic signal characteristics of the strike target according to the noise reduction data, and generate the accuracy index of the strike target according to the noise reduction data;
[0147] The training set generation module 104 is used to generate a training set of a pre-built accuracy analysis model according to the electromagnetic signal characteristics and the accuracy index, and determine the loss function of the pre-built accuracy analysis model, wherein the loss function is:
[0148]
[0149] Wherein, J(Θ) is the loss function value of the pre-built accuracy analysis model, Θ is the model parameter of the pre-built accuracy analysis model, m is the number of training samples in the training set, i is the sample identifier of the training sample in the training set, and y (i) is the sample label of the i-th sample, is the model output representing the i-th sample;
[0150] The model training module 105 is used to perform model training on the pre-built accuracy analysis model using the training set and the loss function to obtain a trained accuracy analysis model;
[0151] The strike strategy generation module 106 is used to determine the optimal electromagnetic emission parameters of the strike target using the trained accuracy analysis model, and generate a strike strategy for the target to be struck according to the optimal electromagnetic emission parameters.
[0152] In the several embodiments provided by the present invention, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.
[0153] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0154] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.
[0155] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.
[0156] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.
Claims
1. A strike accuracy analysis method based on electromagnetic emission, characterized in that: The method comprises: determining electromagnetic emission factors of a strike target, and collecting strike data of the strike target according to the electromagnetic emission factors; Performing data denoising on the strike data to obtain denoised data of the strike data; Extracting electromagnetic signal characteristics of the strike target according to the noise reduction data, and generating an accuracy index of the strike target according to the noise reduction data; A training set of a pre-constructed precision analysis model is generated according to the electromagnetic signal characteristics and the precision index, and a loss function of the pre-constructed precision analysis model is determined, wherein the loss function is: ; in, is the loss function value of the pre-built accuracy analysis model, are the model parameters of the pre-built accuracy analysis model, is the number of training samples in the training set, is the sample identifier of the training sample in the training set, It is The sample labels of samples, It means The model output of samples; Performing model training on the pre-built accuracy analysis model using the training set and the loss function to obtain a trained accuracy analysis model; Determine the optimal electromagnetic emission parameters of the strike target by using the trained precision analysis model, and generate a strike strategy for the target to be struck according to the optimal electromagnetic emission parameters; The electromagnetic emission factors are: electromagnetic signal frequency, electromagnetic signal amplitude and electromagnetic signal phase; The collecting of the strike data of the strike target according to the electromagnetic emission factor comprises: Determining the strike parameters of the strike target according to the electromagnetic emission factors; Performing a target strike on the strike target according to the strike parameter to obtain reflection data of the strike target; The striking data of the striking target is generated according to the striking parameter and the reflection data.
2. The strike accuracy analysis method based on electromagnetic emission according to claim 1, characterized in that: The performing data denoising on the strike data to obtain denoised data of the strike data includes: Performing wavelet transform on the attack data to obtain wavelet coefficients of the attack data; Performing soft threshold processing on the wavelet coefficients according to a preset coefficient threshold to obtain effective coefficients of the wavelet coefficients; Perform an inverse wavelet transform on the effective coefficients to obtain transformation data of the effective coefficients, and determine that the transformation data is noise reduction data of the attack data.
3. The strike accuracy analysis method based on electromagnetic emission according to claim 1, characterized in that: The performing data denoising on the strike data to obtain denoised data of the strike data includes: The adjustment coefficient of the attack data is generated using a preset adjustment coefficient generation algorithm, wherein the preset adjustment coefficient generation algorithm is: ; in, It is The adjustment coefficient vector at time, It is The adjustment coefficient vector at time, is the learning rate, It is The error between the estimated output and the actual output corresponding to the strike data at the time, It is The attack data at the moment, is the time corresponding to the strike data; The striking data is corrected using the adjustment coefficient to obtain corrected data of the striking data, and the corrected data is determined to be noise reduction data of the striking data.
4. The method for analyzing strike accuracy based on electromagnetic emission according to claim 1, characterized in that: The step of extracting the electromagnetic signal characteristics of the strike target according to the noise reduction data comprises: Extracting frequency characteristics of the strike target according to the noise reduction data; generating an amplitude feature of the strike target according to the noise reduction data; Generating a phase feature of the strike target according to the noise reduction data; The electromagnetic signal characteristics of the strike target are generated according to the frequency characteristics, the amplitude characteristics and the phase characteristics.
5. The strike accuracy analysis method based on electromagnetic emission according to claim 1, characterized in that: Generating the accuracy index of the strike target according to the noise reduction data includes: Performing data screening on the noise reduction data to obtain target data of the noise reduction data; Generating a strike evaluation value of the strike target according to the target data, wherein the strike evaluation value includes: hit rate, accuracy and strike deviation; An accuracy index of the strike target is generated according to the strike evaluation value.
6. The method for analyzing strike accuracy based on electromagnetic emission according to claim 1, characterized in that: The step of generating a training set of a pre-built accuracy analysis model according to the electromagnetic signal characteristics and the accuracy index comprises: Establishing a correlation relationship between the electromagnetic signal characteristics and the accuracy index; generating a precision label for the precision indicator; A training set of a pre-built precision analysis model is generated according to the electromagnetic signal characteristics, the association relationship and the precision label.
7. The strike accuracy analysis method based on electromagnetic emission according to any one of claims 1 to 6, characterized in that: The method of using the trained precision analysis model to determine the optimal electromagnetic emission parameters for striking the target includes: Acquiring input features of the strike target; Generate the strike accuracy of the strike target according to the input feature and the trained accuracy analysis model; The optimal electromagnetic emission parameters for striking the target are generated according to the striking accuracy.
8. A strike accuracy analysis device based on electromagnetic emission, characterized in that: The device comprises: A data acquisition module, used to determine the electromagnetic emission factors of the strike target, and to collect strike data of the strike target according to the electromagnetic emission factors; A data denoising module, used for performing data denoising on the attack data to obtain denoised data of the attack data; An accuracy index generating module, used for extracting the electromagnetic signal characteristics of the strike target according to the noise reduction data, and generating the accuracy index of the strike target according to the noise reduction data; A training set generation module is used to generate a training set of a pre-built accuracy analysis model according to the electromagnetic signal characteristics and the accuracy index, and determine a loss function of the pre-built accuracy analysis model, wherein the loss function is: ; in, is the loss function value of the pre-built accuracy analysis model, are the model parameters of the pre-built accuracy analysis model, is the number of training samples in the training set, is the sample identifier of the training sample in the training set, It is The sample labels of samples, It means The model output of samples; A model training module, used to perform model training on the pre-built accuracy analysis model using the training set and the loss function to obtain a trained accuracy analysis model; A strike strategy generation module, used to determine the optimal electromagnetic emission parameters of the strike target using the trained precision analysis model, and generate a strike strategy for the target to be struck according to the optimal electromagnetic emission parameters; The electromagnetic emission factors are: electromagnetic signal frequency, electromagnetic signal amplitude and electromagnetic signal phase; The collecting of the strike data of the strike target according to the electromagnetic emission factor comprises: Determining the strike parameters of the strike target according to the electromagnetic emission factors; Performing a target strike on the strike target according to the strike parameter to obtain reflection data of the strike target; The striking data of the striking target is generated according to the striking parameters and the launching data.
