An electromagnetic pulse identification method and system based on a naive Bayes classification model

Abstract

The application provides an electromagnetic pulse identification method and system based on a Naive Bayes classification model, which comprises the following steps: firstly, simulating NEMP waveforms and LEMP waveforms to obtain each simulation waveform; then, performing normalization processing on the amplitude spectrum of each simulation waveform, and determining a normalized total spectrum according to the distribution frequency band range of the NEMP and the normalization processing result; determining the characteristic value of each simulation waveform according to the normalized total spectrum; obtaining the characteristic values of a preset number of simulation waveforms, and determining the range of the target characteristic value according to the value distribution characteristics of the characteristic values; according to a preset step, respectively calculating the probability of the target characteristic value taking a preset value for the characteristic values of the preset number of simulation waveforms to obtain a first probability value representing the NEMP and a second probability value representing the LEMP; and identifying the electromagnetic pulse according to the size of the first probability value and the second probability value.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of electromagnetic pulse identification, and particularly relates to an electromagnetic pulse identification method and system based on a Naive Bayes classification model. BACKGROUND

[0002] Nuclear electromagnetic pulse (NEMP) is a kind of electromagnetic pulse produced by the interaction of y-ray produced in the moment of nuclear weapon explosion and the surrounding air. NEMP has the characteristics of fast propagation speed, large influence range and long detectable distance, which meets the timeliness requirement of far-zone nuclear explosion detection, so that electromagnetic pulse detection occupies an important position in nuclear explosion detection means.

[0003] However, the detection of NEMP is easily affected by complex electromagnetic environment, and one of the main interference sources is lightning electromagnetic pulse (LEMP). Therefore, the identification of NEMP and LEMP has become the research focus of current far-zone nuclear explosion detection.

[0004] At present, the method for identifying far-zone nuclear explosion and lightning electromagnetic pulse generally starts directly from the time domain and frequency domain characteristics, extracts the time domain or frequency domain characteristics, and then combines the statistical method to distinguish the two signals. In the signal feature extraction process, the traditional method may not fully capture all the key features of the signal, especially in the case of complex changes or noise interference of the signal, which easily leads to incomplete feature extraction, and further affects the accuracy of identification. At the same time, with the increase of propagation distance, the similarity of the two signals will greatly increase, and the identification difficulty will also increase, so the accuracy of the traditional identification method is not high. SUMMARY

[0005] Therefore, the present application provides an electromagnetic pulse identification method and system based on a Naive Bayes classification model, which aims to more effectively identify NEMP and LEMP and improve the identification accuracy.

[0006] The first aspect of the present application provides an electromagnetic pulse identification method based on a Naive Bayes classification model, which comprises:

[0007] Carrying out simulation on NEMP waveforms and LEMP waveforms to obtain each simulation waveform, wherein the simulation waveforms comprise NEMP simulation waveforms and LEMP simulation waveforms;

[0008] Normalizing the amplitude spectrum of each simulation waveform, and determining a normalized total spectrum according to the distribution frequency band range of NEMP and the normalization result;

[0009] Determining the characteristic values of each simulation waveform according to the normalized total spectrum, wherein the characteristic values comprise spectral center frequency, half-peak width, normalized spectrum front average slope, normalized spectrum rear average slope, normalized spectrum skewness and normalized spectrum kurtosis;

[0010] obtaining characteristic values of a preset number of NEMP simulation waveforms and LEMP simulation waveforms, and determining a range of the target characteristic value according to value distribution characteristics of the characteristic values;

[0011] According to the preset step, the probability of the target characteristic value when [a+(i-1)×dx, a+i×dx), i=1, 2,..., n is respectively calculated for the characteristic values of the preset number of NEMP simulation waveforms and LEMP simulation waveforms, to obtain a first probability value representing NEMP and a second probability value representing LEMP, wherein a is the starting value of the range of the target characteristic value, and dx represents the preset step;

[0012] According to the size of the first probability value and the second probability value, the electromagnetic pulse is identified, wherein when the first probability value is greater than the second probability value, the electromagnetic pulse is NEMP; and when the first probability value is less than the second probability value, the electromagnetic pulse is LEMP.

[0013] Further, the NEMP time-domain standard waveform is represented as:

[0014]

[0015] wherein, E(t) represents the electric field intensity at time t, E0 represents the peak field intensity, m1 represents the parameter of the pulse front, n1 represents the parameter of the pulse tail, k1 represents the correction coefficient, and t represents time.

[0016] Further, the LEMP time-domain standard waveform is represented as:

[0017]

[0018] wherein, i(t) represents the current value at time t, I0 represents the current peak value, k represents the correction coefficient, a represents the wave front attenuation coefficient, β represents the wave tail attenuation coefficient, and t represents time.

[0019] Further, the step of simulating the NEMP time-domain standard waveform and the LEMP time-domain standard waveform to obtain each simulation waveform comprises:

[0020] Taking time, position, equivalent, NEMP time-domain standard waveform and LEMP time-domain standard waveform as input parameters, and taking the NEMP waveform and the LEMP waveform of the source area as the reference waveform at 0km, the reference waveform is used to simulate the attenuation of the signal in the propagation process;

[0021] Performing Fourier transform on the NEMP waveform and the LEMP waveform of the source area to obtain the amplitude spectrum and the phase spectrum of the NEMP waveform and the LEMP waveform of the source area;

[0022] Using the very low frequency signal-ionospheric channel propagation model, the amplitude spectrum and the phase spectrum of the NEMP waveform and the LEMP waveform of the source area are analyzed to obtain the amplitude and the phase of different frequency points varying with distance;

[0023] The amplitude and the phase of different frequency points varying with distance are superimposed with the amplitude spectrum and the phase spectrum of the NEMP waveform and the LEMP waveform of the source area to obtain the target amplitude and the target phase of the NEMP waveform and the LEMP waveform at a specified position, and the corresponding target amplitude spectrum and the target phase spectrum are generated;

[0024] The target amplitude spectrum and the target phase spectrum are converted into target time-domain waveforms by inverse Fourier transform to obtain each simulation waveform.

[0025] Further, in the step of normalizing the amplitude spectrum of each simulation waveform, the normalization processing is represented as:

[0026]

[0027] wherein, represents the amplitude spectrum, represents the maximum value in the amplitude spectrum, represents the normalization result, i represents the i th simulation waveform, f represents the frequency, and D represents the propagation distance. Further, in the step of determining the normalized total spectrum according to the distribution frequency band range of the NEMP and the normalization result, the normalized total spectrum is represented as:

[0028]

[0029]

[0030] wherein, represents the normalized total spectrum, the distribution frequency band range of the NEMP is 3 kHz-60 kHz, represents the normalization result.

[0031] Further, in the step of determining the characteristic value of each simulation waveform according to the normalized total spectrum, the spectral center frequency is represented as:

[0032]

[0033] represents the spectral center frequency;

[0034] The half-peak width is represented as:

[0035]

[0036] represents the half-peak width, is ​inverse function of the normalized spectrum, m represents the mth maximum value in the normalized spectrum;

[0037] The normalized spectrum front average slope is represented as:

[0038]

[0039] The normalized spectrum front average slope is represented as: the second maximum value greater than 0.5 from 3kHz to 60kHz in the normalized spectrum;

[0040] The normalized spectrum back average slope is represented as:

[0041]

[0042] The normalized spectrum back average slope is represented as: the second maximum value greater than 0.5 from 3kHz to 60kHz in the normalized spectrum;

[0043] The normalized spectrum skewness is represented as:

[0044]

[0045] The normalized spectrum skewness is represented as:

[0046] The normalized spectrum kurtosis is represented as:

[0047]

[0048] The normalized spectrum kurtosis is represented as:

[0049] The second aspect of the embodiment of the application provides an electromagnetic pulse recognition system based on a naive Bayes classification model, which is used to realize the electromagnetic pulse recognition method based on the naive Bayes classification model provided in the first aspect, and the system comprises:

[0050] The simulation module is used to simulate NEMP waveforms and LEMP waveforms to obtain simulation waveforms, wherein the simulation waveforms comprise NEMP simulation waveforms and LEMP simulation waveforms.

[0051] The normalization processing module is used to normalize the amplitude spectrum of each simulation waveform, and determine a normalized total spectrum according to the distribution frequency band range of the NEMP and the normalization processing result.

[0052] The characteristic value determination module is used to determine characteristic values of each simulation waveform according to the normalized total spectrum, wherein the characteristic values comprise a spectrum center frequency, a half-peak width, a normalized spectrum front average slope, a normalized spectrum back average slope, a normalized spectrum skewness and a normalized spectrum kurtosis.​​

[0053] The acquisition module is configured to acquire characteristic values of a preset number of NEMP simulation waveforms and LEMP simulation waveforms, and determine a range of the target characteristic value according to a value distribution characteristic of the characteristic values.

[0054] The probability and statistics module is configured to respectively statistically calculate probabilities of the target characteristic value in [a+(i-1)×dx, a+i×dx), i=1, 2,..., n for the characteristic values of the preset number of NEMP simulation waveforms and LEMP simulation waveforms according to a preset step length, so as to obtain a first probability value representing NEMP and a second probability value representing LEMP, wherein a is a starting value of the range of the target characteristic value, and dx represents the preset step length.

[0055] The identification module is configured to identify the electromagnetic pulse according to the sizes of the first probability value and the second probability value, wherein when the first probability value is greater than the second probability value, the electromagnetic pulse is NEMP; and when the first probability value is less than the second probability value, the electromagnetic pulse is LEMP.

[0056] A third aspect of the embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement the electromagnetic pulse identification method based on the Naive Bayes classification model provided in the first aspect.

[0057] A fourth aspect of the embodiment of the present application provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the electromagnetic pulse identification method based on the Naive Bayes classification model provided in the first aspect when executing the program.

[0058] The method provided in the embodiment of the present application comprises the following steps: simulating NEMP time domain standard waveforms and LEMP time domain standard waveforms to obtain simulation waveforms; performing normalization processing on amplitude spectrums of the simulation waveforms, and determining a normalized total spectrum according to a distribution frequency band range of the NEMP and the normalized processing result; determining characteristic values of the simulation waveforms according to the normalized total spectrum; obtaining characteristic values of a preset number of NEMP simulation waveforms and LEMP simulation waveforms, selecting a spectral center frequency as a target characteristic value, and determining a range of the target characteristic value according to a frequency range of the electromagnetic pulse signal; according to a preset step length, respectively calculating probabilities of the target characteristic value taking [a+(i-1)×dx, a+i×dx), i=1, 2,..., n, to obtain a first probability value representing the NEMP and a second probability value representing the LEMP, wherein a is a starting value of the range of the target characteristic value, and dx represents the preset step length; and identifying the electromagnetic pulse according to the sizes of the first probability value and the second probability value. BRIEF DESCRIPTION OF DRAWINGS

[0059] Figure 1 An implementation flowchart of the electromagnetic pulse identification method based on the Naive Bayes classification model is provided for the first embodiment of the present application.

[0060] Figure 2 A structural block diagram of the electromagnetic pulse identification system based on the Naive Bayes classification model is provided for the second embodiment of the present application.

[0061] Figure 3 A structural block diagram of the electronic device is provided for the third embodiment of the present application. DETAILED DESCRIPTION

[0062] In order to make the present application more comprehensively understood, the present application will be described more fully below with reference to the accompanying drawings. The present application is shown in several embodiments in the drawings. However, the present application can be realized in many different forms and is not limited to the embodiments described herein. On the contrary, these embodiments are provided to make the disclosure of the present application more thorough and comprehensive.

[0063] It should be noted that when an element is referred to as being "fixedly attached" to another element, it can be directly on the other element or there can be an intervening element. When an element is referred to as being "connected" to another element, it can be directly connected to the other element or intervening elements can be present. The terms "vertical", "horizontal", "left", "right", and the like as used herein are for purposes of illustration and description only.

[0064] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the description herein is for describing particular embodiments only and is not intended to be limiting of the application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.

[0065] Embodiment one

[0066] According to the embodiment of the present application, there is provided an electromagnetic pulse identification method based on a Naive Bayes classification model. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown herein.

[0067] In this embodiment one, there is provided an electromagnetic pulse identification method based on a Naive Bayes classification model, which can be used in electronic devices such as computers. Please refer to Figure 1 , Figure 1 The implementation flowchart of the electromagnetic pulse identification method based on a Naive Bayes classification model provided by the embodiment one of the present application is shown, which specifically includes steps S01 to S06.

[0068] Step S01, NEMP waveform and LEMP waveform are simulated to obtain each simulation waveform, and the simulation waveform includes NEMP simulation waveform and LEMP simulation waveform.

[0069] Specifically, the NEMP time-domain standard waveform is represented as:

[0070]

[0071] wherein, E(t) represents the electric field intensity at time t, and the unit is volt / meter, E0 represents the peak field intensity, m1 represents the parameter of the pulse front, n1 represents the parameter of the pulse tail, k1 represents the correction coefficient, and t represents the time, and the unit is second. In this embodiment, E0=5000v / m, m1=4×10 7 s -1 , n1=6×10 8 s -1 , k1=1.3;

[0072] The LEMP time-domain standard waveform is represented as:

[0073]

[0074] Wherein, i(t) represents the current value at t time, I0 represents the current peak value, k represents the correction coefficient, a represents the wave front attenuation coefficient, b represents the wave tail attenuation coefficient, and t represents time. In the embodiment, a is 7.714*10 4 , b is 2.489*10 5 , and k is 2.33.

[0075] It should be noted that, in order to establish a sample library, that is, a database composed of simulation waveforms, firstly, time, position, equivalent, NEMP time domain standard waveform and LEMP time domain standard waveform are taken as input parameters, and the NEMP waveform and the LEMP waveform of the source area are taken as reference waveforms at 0km, so as to simulate the attenuation of signals in the propagation process by using the reference waveforms;

[0076] The NEMP waveform and the LEMP waveform of the source area are subjected to Fourier transform, so as to obtain the amplitude spectrum and the phase spectrum of the NEMP waveform and the LEMP waveform of the source area;

[0077] The amplitude spectrum and the phase spectrum of the NEMP waveform and the LEMP waveform of the source area are analyzed by using a very low frequency signal ground-ionosphere channel propagation model, so as to obtain the results of the amplitude and the phase of different frequency points changing with distance. It should be noted that the earth surface and the low ionosphere have excellent reflection performance on the very low frequency electromagnetic wave, the distance between the low ionosphere and the ground is close to the wavelength of the very low frequency electromagnetic wave, and the very low frequency electromagnetic wave propagates in the space composed of the earth and the ionosphere like in a waveguide. The very low frequency ground-ionosphere channel propagation model is a model for researching the propagation characteristics of the very low frequency electromagnetic wave in the waveguide structure formed between the earth surface and the ionosphere.

[0078] The results of the amplitude and the phase of different frequency points changing with distance are superimposed with the amplitude spectrum and the phase spectrum of the NEMP waveform and the LEMP waveform of the source area, so as to obtain the target amplitude and the target phase of the NEMP waveform and the LEMP waveform at a specified position, and to generate the corresponding target amplitude spectrum and target phase spectrum.

[0079] The target amplitude spectrum and the target phase spectrum are converted into target time domain waveforms by inverse Fourier transform, so as to obtain each simulation waveform.

[0080] In some other embodiments of the application, in order to make the simulation waveforms cover different conditions as much as possible, the source points and the receiving points can be selected uniformly in space, the dates can be selected to cover different seasons, and the local times can be selected to cover different typical moments in the day and night. This is done to ensure that the simulation samples have good spatiotemporal coverage.

[0081] Step S02, the amplitude spectrum of each simulation waveform is subjected to normalization processing, and the normalized total spectrum is determined according to the distribution frequency band range of the NEMP and the normalization processing result.

[0082] Specifically, if E i (t, D) represents the waveform of the i-th sample, the amplitude spectrum can be represented as (f, D), which is normalized and represented as:

[0083]

[0084] wherein, represents the amplitude spectrum, represents the maximum value in represents the normalized result, i represents the i-th simulated waveform, f represents the frequency, and D represents the propagation distance.

[0085] Since the main components of the nuclear electromagnetic pulse waveform at a long distance are concentrated in the 0 kHz~60 kHz frequency band, the following characteristic quantities are defined in this frequency band.

[0086] The normalized total spectrum is represented as:

[0087]

[0088] wherein, represents the normalized total spectrum, the distribution frequency band of the NEMP is 3 kHz~60 kHz, represents the normalized result.

[0089] Step S03, according to the normalized total spectrum, determining the characteristic values of each simulated waveform, the characteristic values including the spectral center frequency, the half-peak width, the normalized spectrum front average slope, the normalized spectrum rear average slope, the normalized spectrum skewness, and the normalized spectrum kurtosis.

[0090] Specifically, the spectral center frequency is represented as:

[0091]

[0092] represents the spectral center frequency;

[0093] The half-peak width is represented as:

[0094]

[0095] represents the half-peak width, is the inverse function of , m represents the m-th maximum value in If the sum symbol is not 1 when the maximum value in , it should be the sum of the half-peak widths corresponding to the maximum values. In particular, the normalized spectrum can be written as:

[0096]

[0097] The number of maxima greater than half the peak:

[0098]

[0099] When n i When n is 1 or 2, the average slope of the leading edge and the trailing edge can be defined. When n is greater than 2, the definition is not valid.

[0100] The normalized spectrum leading edge average slope is represented as:

[0101]

[0102] The normalized spectrum leading edge average slope is represented as: The normalized spectrum leading edge average slope is represented as: The first maximum greater than 0.5 from 3 kHz to 60 kHz in the number;

[0103] The normalized spectrum trailing edge average slope is represented as:

[0104]

[0105] The normalized spectrum trailing edge average slope is represented as: The normalized spectrum trailing edge average slope is represented as: The second maximum greater than 0.5 from 3 kHz to 60 kHz in the number;

[0106] The normalized spectrum skewness is represented as:

[0107]

[0108] The normalized spectrum skewness is represented as:

[0109] The normalized spectrum kurtosis is represented as:

[0110]

[0111] The normalized spectrum kurtosis is represented as:

[0112] Step S04, obtaining the characteristic values of a preset number of NEMP simulation waveforms and LEMP simulation waveforms, and determining the range of the target characteristic value according to the value distribution characteristics of the characteristic values.

[0113] Among them, the characteristic values of a preset number of NEMP simulation waveforms and LEMP simulation waveforms are obtained, the spectral center frequency is selected as the target characteristic value, and the range of the target characteristic value is determined according to the frequency range of the electromagnetic pulse signal.

[0114] In the embodiment, 10000000 NEMP samples are obtained and recorded as ω1, and 10000000 LEMP samples are obtained and recorded as ω2, wherein the samples are simulation waveforms, and spectrum center frequency data of each NEMP and LEMP sample is extracted as x. According to the frequency range of the electromagnetic pulse signal, the range of x is limited to 3 kHz-60 kHz.

[0115] In step S05, according to a preset step, the probability of the target feature value being in [a+(i-1)×dx, a+i×dx), i=1, 2,..., n is obtained for the feature values of a preset number of NEMP simulation waveforms and LEMP simulation waveforms, to obtain a first probability value representing NEMP and a second probability value representing LEMP, wherein a is a starting value of the range of the target feature value, and dx represents the preset step.

[0116] Taking the spectrum center frequency as an example, the frequency range is 3-60 kHz, the step dx can be set to 0.1 kHz, and the spectrum center frequency feature value statistical interval is [3+(i-1)×dx, 3+i×dx), i=1, 2,..., 570.

[0117] In step S06, according to the sizes of the first probability value and the second probability value, the electromagnetic pulse is identified, wherein when the first probability value is greater than the second probability value, the electromagnetic pulse is NEMP; and when the first probability value is less than the second probability value, the electromagnetic pulse is LEMP.

[0118] It should be noted that assuming that there is an unknown sample, first, the spectrum center frequency x of the sample is extracted, and the likelihood ratio of the sample can be obtained by substituting x into p(x|ω1) and p(x|ω2) respectively:

[0119]

[0120] According to the number of two types of samples used to obtain the curves of p(x|ω1) and p(x|ω2), the likelihood ratio threshold value can be directly calculated:

[0121]

[0122] Since the number of NEMP samples and the number of LEMP samples are both 10000000, the threshold value is At this time, the threshold value is the position of the intersection of the two curves, and the minimum error rate Bayesian decision can be completed by directly comparing the sizes of p(x|ω1) and p(x|ω2), and the type of which is larger is determined.

[0123] To evaluate the performance of the above decision method, first calculate the first and second type error rates, and draw the ROC curve, specifically, the first type error rate is the false positive rate, i.e. the probability of misreporting a non-nuclear explosion sample as a nuclear explosion event, also known as the false alarm rate or false positive rate, denoted as:

[0124]

[0125] That is, the error rate of deciding the second type sample as the first type.

[0126] The second type error rate is the false negative rate, i.e. the probability of misreporting a nuclear explosion sample as a non-nuclear explosion event, also known as the false negative rate, denoted as:

[0127]

[0128] That is, the error rate of deciding the first type sample as the second type.

[0129] Since the threshold has been determined, the boundaries of R1 and R2 are also determined, and the calculation method of the two types of error rates can be directly obtained from p(x|ω1) and p(x|ω2), because there is the following relationship between them:

[0130]

[0131] In actual calculation, since p(x|ω1) and p(x|ω2) give discrete values, the sum can be directly calculated with the boundary points. Note that the vertical axis of the ROC curve is the sensitivity Sn, i.e. the true positive rate, Sn=1-P1(e); the horizontal axis is the false positive rate, P2(e).

[0132] By controlling the prior probability, i.e. the number of two types of samples, the threshold size can be adjusted. In one extreme case, all samples are negative samples, i.e. LEMP samples, at this time p(x|ω1)=0, at this time any unknown sample will be judged as negative sample, the false positive rate is 0, the true positive rate is also 0, in the ROC curve gets (0, 0) point. In another extreme case, all samples are positive samples, i.e. NEMP samples, at this time p(x|ω2)=0, at this time any unknown sample will be judged as positive sample, the true positive rate is 1, the false positive rate is also 1, in the ROC curve gets (1, 1) point.

[0133] N1 and N2 represent the sample number of NEMP and LEMP respectively, the total sample number is controlled at 1 million, and the proportion of N1 and N2 is gradually adjusted to change the threshold size. For example, with a step size of 10,000, the proportions are set to 0:100, 1:99, 2:98, 3:97,..., 99:1, and 100:0, respectively. The above settings will form a total of 101 prior probabilities, corresponding to 101 thresholds, and each threshold can obtain the error rate of two categories, and then the ROC curve can be drawn.

[0134] Further, for a classifier composed of a single feature parameter, the probability of correct identification of NEMP by the minimum error rate Bayesian decision is P (i.e. the recognition rate is P), and if n stations simultaneously collect the event, there are n waveforms. Since they are the same parameters, their monitoring performance is the same, and the theoretically correct identification rate of the electromagnetic pulse through voting fusion is:

[0135] P=P n +P n-1 (1-P)×n+...+P n / 2 (1-P) n-n / 2 ×n

[0136] Calculation example:

[0137] Assuming that the recognition rates of 4 feature parameters are 99%, 98%, 95%, and 90% respectively when 5 monitoring stations are used, the recognition rate obtained by multi-parameter voting fusion is:

[0138] The probability of correct identification of 4 feature parameters is:

[0139] 99%×98%×95%×90%=82.9521%;

[0140] The probability of correct identification of 3 feature parameters is:

[0141] 99%×98%×95%×10%+99%×98%×5%×90%+99%×2%×95%×90%+1%×98%×95%×90%=9.2169%+4.3659%+1.6929%+0.8379%=16.1136%;

[0142] The probability of correct identification of 2 feature parameters is:

[0143] 99%×98%×5%×10%+99%×2%×5%×90%+1%×2%×95%×90%+1%×98%×95%×10%=0.4851%+0.0891%+0.0171%+0.0931%=0.6844%;

[0144] The comprehensive recognition rate is: 99.0657%+0.6844%=99.7501%.

[0145] To sum up, the electromagnetic pulse recognition method based on the Naive Bayes classification model in the above-mentioned embodiments of the application, the method simulates the NEMP time domain standard waveform and the LEMP time domain standard waveform to obtain each simulation waveform; the amplitude spectrum of each simulation waveform is normalized, and the normalized total spectrum is determined according to the distribution frequency band range of the NEMP and the normalization result; the characteristic value of each simulation waveform is determined according to the normalized total spectrum; the characteristic values of a preset number of simulation waveforms are obtained, the spectral center frequency is selected as the target characteristic value, and the range of the target characteristic value is determined according to the frequency range of the electromagnetic pulse signal; the probability of the target characteristic value taking a preset value is respectively calculated according to the preset step length for the characteristic values of the preset number of simulation waveforms, and the first probability value representing the NEMP and the second probability value representing the LEMP are obtained; the electromagnetic pulse is recognized according to the size of the first probability value and the second probability value.

[0146] Embodiment two

[0147] Please refer to Figure 2 , Figure 2 is a structure block diagram of an electromagnetic pulse recognition system based on a Naive Bayes classification model provided by the second embodiment of the application, and the electromagnetic pulse recognition system based on the Naive Bayes classification model 200 is used to realize the above-mentioned embodiments and preferred embodiments, and will not be described again. As used below, the term "module" can be a combination of software and / or hardware that realizes a predetermined function. Although the devices described in the following embodiments are preferably realized in software, the realization of hardware, or a combination of software and hardware, is also possible and conceived.

[0148] Specifically, the electromagnetic pulse recognition system based on the Naive Bayes classification model 200 includes a simulation module 21, a normalization processing module 22, a characteristic value determination module 23, an acquisition module 24, a probability statistical module 25, and a recognition module 26, wherein:

[0149] The simulation module 21 is used to simulate NEMP waveforms and LEMP waveforms to obtain each simulation waveform, and the simulation waveform includes NEMP simulation waveforms and LEMP simulation waveforms, and the NEMP time domain standard waveform is represented as:

[0150]

[0151] wherein, represents the electric field intensity at time t, represents the peak field intensity, m1 represents the parameter of the pulse front, n1 represents the parameter of the pulse tail, k1 represents the correction coefficient, and t represents time, and the LEMP time domain standard waveform is represented as:

[0152]

[0153] wherein i(t) represents the current value at time t, I0 represents the current peak value, k represents the correction coefficient, a represents the wave front attenuation coefficient, b represents the wave tail attenuation coefficient, and t represents time;

[0154] The normalization processing module 22 is configured to normalize the amplitude spectrum of each simulation waveform, and determine a normalized total spectrum according to the distribution frequency band range of the NEMP and the normalization processing result, wherein the normalization processing is represented as:

[0155]

[0156] wherein, represents the amplitude spectrum, represents the maximum value in the amplitude spectrum, represents the normalization result, i represents the i-th simulation waveform, f represents the frequency, and D represents the propagation distance, and the normalized total spectrum is represented as:

[0157]

[0158] wherein, represents the normalized total spectrum, and the distribution frequency band range of the NEMP is 3 kHz-60 kHz, represents the normalization result;

[0159] The characteristic value determination module 23 is configured to determine the characteristic values of each simulation waveform according to the normalized total spectrum, wherein the characteristic values include the spectral center frequency, the half-peak width, the normalized spectrum front average slope, the normalized spectrum tail average slope, the normalized spectrum skewness, and the normalized spectrum kurtosis, and the spectral center frequency is represented as:

[0160]

[0161] represents the spectral center frequency;

[0162] The half-peak width is represented as:

[0163]

[0164] represents the half-peak width, is the inverse function of , and m represents the m-th maximum value in ;

[0165] The normalized spectrum front average slope is represented as:

[0166] ​

[0167] denotes the normalized spectrum front average slope, denotes the normalized spectrum front average slope, the first maximum value greater than 0.5 in the range of 3 kHz to 60 kHz;

[0168] The normalized spectrum tail average slope is denoted as:

[0169]

[0170] denotes the normalized spectrum tail average slope, denotes the normalized spectrum tail average slope, the second maximum value greater than 0.5 in the range of 3 kHz to 60 kHz;

[0171] The normalized spectrum skewness is denoted as:

[0172]

[0173] L denotes the normalized spectrum skewness;

[0174] The normalized spectrum kurtosis is denoted as:

[0175]

[0176] V denotes the normalized spectrum kurtosis;

[0177] The acquisition module 24 is configured to acquire characteristic values of a preset number of NEMP simulation waveforms and LEMP simulation waveforms, and determine a range of the target characteristic value according to a value distribution feature of the characteristic values.

[0178] The probability and statistics module 25 is configured to respectively statistically calculate probabilities of the target characteristic value taking [a+(i-1)×dx, a+i×dx), i=1, 2,..., n according to a preset step length for the characteristic values of the preset number of NEMP simulation waveforms and LEMP simulation waveforms, so as to obtain a first probability value representing NEMP and a second probability value representing LEMP, wherein a is a starting value of the range of the target characteristic value, and dx represents the preset step length.

[0179] The identification module 26 is configured to identify the electromagnetic pulse according to sizes of the first probability value and the second probability value, wherein when the first probability value is greater than the second probability value, the electromagnetic pulse is NEMP; and when the first probability value is less than the second probability value, the electromagnetic pulse is LEMP.

[0180] Further, in some optional embodiments of the present application, the simulation module 21 comprises:

[0181] A definition unit is configured to take time, location, equivalent, NEMP time-domain standard waveform and LEMP time-domain standard waveform as input parameters, and take NEMP waveform and LEMP waveform of a source area as reference waveforms at 0 km, so as to simulate the attenuation of signals in the propagation process by using the reference waveforms.

[0182] A Fourier transform unit is configured to perform Fourier transform on the NEMP waveform and LEMP waveform of the source area, so as to obtain amplitude spectrum and phase spectrum of the NEMP waveform and LEMP waveform of the source area.

[0183] An analysis unit is configured to analyze the amplitude spectrum and phase spectrum of the NEMP waveform and LEMP waveform of the source area by using a very low frequency signal-ionosphere channel propagation model, so as to obtain the results of amplitude and phase changes with distance at different frequency points.

[0184] A superposition unit is configured to superimpose the results of amplitude and phase changes with distance at different frequency points and the amplitude spectrum and phase spectrum of the NEMP waveform and LEMP waveform of the source area, so as to obtain target amplitude and target phase of the NEMP waveform and LEMP waveform at a specified location, and generate corresponding target amplitude spectrum and target phase spectrum.

[0185] An inverse Fourier transform unit is configured to convert the target amplitude spectrum and target phase spectrum into target time-domain waveforms by inverse Fourier transform, so as to obtain each simulation waveform.

[0186] Embodiment Three

[0187] Another aspect of the present application also provides an electronic device, please refer to Figure 3 , which is an electronic device in the embodiment three of the present application, comprising a memory 20, a processor 10 and a computer program 30 stored in the memory and executable on the processor, wherein the processor 10 implements the above-mentioned electromagnetic pulse recognition method based on the Naive Bayes classification model when executing the computer program 30.

[0188] In some embodiments, the processor 10 can be a central processing unit (CPU), a controller, a microcontroller, a microprocessor or other data processing chip, which is used to run program codes or process data stored in the memory 20, such as executing access restriction programs.

[0189] The memory 20 includes at least one type of readable storage medium, such as a flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. The memory 20 can be an internal storage unit of the electronic device in some embodiments, such as a hard disk of the electronic device. The memory 20 can also be an external storage device of the electronic device in other embodiments, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device. Further, the memory 20 can include both the internal storage unit and the external storage device of the electronic device. The memory 20 can be used not only to store application software and various data of the electronic device, but also to temporarily store data that has been output or will be output.

[0190] It should be noted that, Figure 3 The illustrated structure does not limit the electronic device, and in other embodiments, the electronic device can include fewer or more components than illustrated, or combine certain components, or arrange different components.

[0191] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program. The program is executed by a processor to implement the method for identifying electromagnetic pulses based on a naive Bayes classification model.

[0192] Those skilled in the art can understand that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a list of executable instructions for implementing the logic function, which can be specifically implemented in any computer readable medium for use by or in conjunction with an instruction execution system, device or apparatus, such as a computer-based system, a system including a processor or other system that can fetch and execute instructions from the instruction execution system, device or apparatus. For the present specification, the "computer readable medium" can be any device that can contain, store, communicate, propagate or transport programs for use by or in conjunction with the instruction execution system, device or apparatus, or in conjunction with these instruction execution systems, devices or apparatus.

[0193] More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection (electronic) having one or more wires, a portable computer diskette (magnetic), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can also be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, for example via an optical scanner, then compiled, interpreted, or otherwise processed, and stored in a computer memory in a form that is then suitable for use by the computer. Program code embodied on a computer-readable medium can be transmitted using any apparatus adapted to transmit such a program code, including a modem, or other fetal communication apparatus.

[0194] It should be understood that aspects of the application can be implemented in hardware, software, firmware or combinations thereof. In the embodiments described above, various steps or methods can be implemented, for example, by software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, and in another embodiment, any of the following techniques, which are well known in the art, can be used to implement the application: a hybrid of the techniques mentioned above; a combination of one or more of the techniques mentioned above; or one or more other techniques suitable for use in the computer-based systems described above.

[0195] In the description of the specification, the description of the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the application. In the specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in one or more embodiments or examples.

[0196] The above embodiments only express several implementation manners of the application, which are described in detail and specifically, but cannot be understood as limiting the scope of the patent of the application. It should be noted that for those skilled in the art, several modifications and improvements can be made without departing from the concept of the application, which are all within the protection scope of the application. Therefore, the protection scope of the patent of the application should be subject to the appended claims.

Claims

1. A method for electromagnetic pulse recognition based on a Naive Bayes classification model, characterized in that, The method comprises: Carrying out simulation on NEMP waveforms and LEMP waveforms to obtain simulation waveforms, wherein the simulation waveforms comprise NEMP simulation waveforms and LEMP simulation waveforms; Normalizing the amplitude spectrum of each simulation waveform, and determining a normalized total spectrum according to the distribution frequency band range of the NEMP and the normalized processing result; Determining characteristic values of each simulation waveform according to the normalized total spectrum, wherein the characteristic values comprise a spectral center frequency, a half-peak width, a normalized spectrum front average slope, a normalized spectrum rear average slope, a normalized spectrum skewness, and a normalized spectrum kurtosis; Obtaining characteristic values of a preset number of NEMP simulation waveforms and LEMP simulation waveforms, and determining a range of a target characteristic value according to the value distribution characteristics of the characteristic values; According to a preset step, the probability of the target characteristic value being [a+(i-1)×dx, a+i×dx), i=1, 2,..., n is calculated for the characteristic values of the preset number of NEMP simulation waveforms and LEMP simulation waveforms, to obtain a first probability value representing the NEMP and a second probability value representing the LEMP, wherein a is a starting value of the range of the target characteristic value, and dx represents the preset step; According to the sizes of the first probability value and the second probability value, the electromagnetic pulse is identified, wherein when the first probability value is greater than the second probability value, the electromagnetic pulse is the NEMP; and when the first probability value is less than the second probability value, the electromagnetic pulse is the LEMP; The step of carrying out simulation on NEMP waveforms and LEMP waveforms to obtain simulation waveforms comprises: Taking time, position, equivalent, NEMP time-domain standard waveform, and LEMP time-domain standard waveform as input parameters, and taking the NEMP waveform and the LEMP waveform of the source area as reference waveforms at 0 km, to simulate the attenuation of signals in the propagation process by using the reference waveforms; Performing Fourier transform on the NEMP waveform and the LEMP waveform of the source area to obtain the amplitude spectrum and the phase spectrum of the NEMP waveform and the LEMP waveform of the source area; Using a very low frequency signal-ground-ionosphere channel propagation model, analyzing the amplitude spectrum and the phase spectrum of the NEMP waveform and the LEMP waveform of the source area to obtain the results of the amplitude and the phase varying with distance at different frequency points; Superimposing the results of the amplitude and the phase varying with distance at different frequency points on the amplitude spectrum and the phase spectrum of the NEMP waveform and the LEMP waveform of the source area to obtain the target amplitude and the target phase of the NEMP waveform and the LEMP waveform at a specified position, and generating corresponding target amplitude spectrum and target phase spectrum; Converting the target amplitude spectrum and the target phase spectrum into target time-domain waveforms by inverse Fourier transform to obtain simulation waveforms. 2.The method of claim 1, wherein, The NEMP time-domain standard waveform is represented as: wherein E(t) represents the electric field strength at time t, Epkrepresents the peak field strength, mlrepresents a parameter of the pulse front, nlrepresents a parameter of the pulse back, klrepresents a correction factor, and t represents time. 3.The method of claim 2, wherein, The LEMP time-domain standard waveform is represented as: Wherein, i(t) represents the current value at time t, I0 represents the peak current value, k represents the correction coefficient, a represents the wave front attenuation coefficient, b represents the wave tail attenuation coefficient, and t represents the time. 4.The method of claim 3, wherein, In the step of normalizing the amplitude spectrum of each simulation waveform, the normalization processing is represented as: wherein, denotes the amplitude spectrum, denotes the maximum in denotes the normalized result, i denotes the i-th simulated waveform, f denotes the frequency, and D denotes the propagation distance. 5.The method of claim 4, wherein, The normalized total spectrum is expressed as: wherein, represents the normalized total spectrum, the distribution band range of NEMP is 3 kHz~60 kHz, represents the normalized result. 6.The method of claim 5, wherein, The spectral center frequency is expressed as: denotes the spectral center frequency; The half-peak width is expressed as: denotes the half-peak width, is the inverse function of m denotes the mth maximum in the sequence The normalized spectrum front average slope is expressed as: represents the average slope of the leading edge of the normalized spectrum, represents the first maximum greater than 0.5 from 3 kHz to 60 kHz; The normalized spectrum rear average slope is expressed as: represents the average slope of the trailing edge of the normalized spectrum, represents the second maximum greater than 0.5 from 3 kHz to 60 kHz; The normalized spectrum skewness is expressed as: L represents the normalized spectrum skewness. The normalized spectrum kurtosis is expressed as: V represents the normalized spectrum kurtosis.

7. An electromagnetic pulse recognition system based on a Naive Bayes classification model, characterized by, The system for implementing the electromagnetic pulse identification method based on the Naive Bayes classification model according to any one of claims 1-6 comprises: A simulation module for simulating NEMP waveforms and LEMP waveforms to obtain simulation waveforms, wherein the simulation waveforms comprise NEMP simulation waveforms and LEMP simulation waveforms; A normalization processing module for normalizing the amplitude spectrum of each simulation waveform and determining a normalized total spectrum according to the distribution frequency band range of NEMP and the normalization processing result; A feature value determination module for determining feature values of each simulation waveform according to the normalized total spectrum, wherein the feature values comprise a spectral center frequency, a half-peak width, a normalized spectrum front average slope, a normalized spectrum rear average slope, a normalized spectrum skewness, and a normalized spectrum kurtosis; An acquisition module for acquiring feature values of a preset number of NEMP simulation waveforms and LEMP simulation waveforms and determining a range of a target feature value according to the feature value distribution characteristics; A probability and statistics module for respectively calculating the probability of the target feature value taking [a + (i-1) × dx, a + i × dx), i = 1, 2,..., n according to a preset step size for the feature values of the preset number of NEMP simulation waveforms and LEMP simulation waveforms, to obtain a first probability value representing NEMP and a second probability value representing LEMP, wherein a is a starting value of the range of the target feature value, and dx represents the preset step size; An identification module for identifying an electromagnetic pulse according to the sizes of the first probability value and the second probability value, wherein when the first probability value is greater than the second probability value, the electromagnetic pulse is NEMP; and when the first probability value is less than the second probability value, the electromagnetic pulse is LEMP.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the electromagnetic pulse identification method based on the Naive Bayes classification model according to any one of claims 1-6.

9. An electronic device, comprising: The computer program is stored in the memory and executable on the processor, and the processor implements the electromagnetic pulse identification method based on the Naive Bayes classification model according to any one of claims 1-6 when executing the program.

Images