Electromagnetic pulse identification method and system based on naive Bayes classification model

Abstract

The invention provides an electromagnetic pulse identification method and system based on a naive Bayes classification model, and the method comprises the steps: firstly carrying out the simulation of NEMP waveforms and LEMP waveforms, and obtaining each simulation waveform; performing normalization processing on the amplitude spectrum of each simulation waveform, and determining a normalized total spectrum according to a distribution frequency band range of the NEMP and a normalization processing result; determining a characteristic value of each simulation waveform according to the normalized total spectrum; obtaining feature values of a preset number of simulation waveforms, and determining the range of the target feature value according to the value distribution characteristics of the feature values; according to a preset step length, the probability of taking a preset value of the target feature value is counted for the feature values of the preset number of simulation waveforms, and a first probability value representing the NEMP and a second probability value representing the LEMP are obtained; and identifying the electromagnetic pulse according to the first probability value and the second probability value.

Description

Technical Field

[0001] The invention belongs to the technical field of electromagnetic pulse identification, and in particular relates to an electromagnetic pulse identification method and system based on a naive Bayesian classification model. Background Art

[0002] Nuclear explosion electromagnetic pulse (NEMP) is an electromagnetic pulse generated by the interaction of gamma rays produced by a nuclear weapon explosion with the surrounding air. NEMP has the characteristics of fast propagation speed, large impact range and long detection distance, which meets the timeliness requirements of remote nuclear explosion detection, making electromagnetic pulse detection occupy an important position in nuclear explosion detection methods.

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

[0004] At present, the method for identifying distant nuclear explosions and lightning electromagnetic pulses generally starts directly from the characteristics of the time domain and frequency domain, extracts the characteristics of the time domain or frequency domain, and then combines statistical methods to distinguish the two signals. In the process of signal feature extraction, traditional methods may not be able to fully capture all the key features of the signal, especially when the signal has complex changes or noise interference, which can easily lead to incomplete feature extraction and affect the accuracy of recognition. At the same time, as the propagation distance increases, the similarity of the two signals will increase significantly, and the difficulty of recognition will increase accordingly. The accuracy of traditional recognition methods is not high. Summary of the invention

[0005] Based on this, an embodiment of the present invention provides an electromagnetic pulse identification method and system based on a naive Bayes classification model, aiming to more effectively identify NEMP and LEMP and improve the identification accuracy.

[0006] A first aspect of an embodiment of the present invention provides an electromagnetic pulse identification method based on a naive Bayes classification model, the method comprising: Conducting simulations of NEMP waveforms and LEMP waveforms to obtain various simulation waveforms, wherein the simulation waveforms include NEMP simulation waveforms and LEMP simulation waveforms; The amplitude spectrum of each simulation waveform is normalized, and the normalized total spectrum is determined according to the distribution frequency band range of NEMP and the normalization processing result; According to the normalized total spectrum, the characteristic values ​​of each simulation waveform are determined, wherein the characteristic values ​​include the spectrum center frequency, the half-peak width, the average slope of the normalized spectrum front edge, the average slope of the normalized spectrum rear edge, the normalized spectrum skew and the normalized spectrum kurtosis; Acquire a preset number of NEMP simulation waveforms and LEMP simulation waveforms, and determine the range of the target characteristic value according to the characteristic value distribution characteristics; According to a preset step size, the probability of the target characteristic value taking [a+(i-1)×dx, a+i×dx), i=1, 2, ..., n, is counted for the characteristic values ​​of a preset number of NEMP simulation waveforms and LEMP simulation waveforms respectively, to obtain a first probability value representing NEMP and a second probability value representing LEMP, wherein a is the starting value of the target characteristic value range, and dx represents the preset step size; The electromagnetic pulse is identified according to the magnitude 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; when the first probability value is less than the second probability value, the electromagnetic pulse is LEMP.

[0007] Furthermore, the NEMP time domain standard waveform is expressed as:

[0008] in, represents the electric field strength at time t, represents the peak field strength, m1 represents the parameters of the pulse leading edge, n1 represents the parameters of the pulse trailing edge, k1 represents the correction coefficient, and t represents time.

[0009] Furthermore, the LEMP time domain standard waveform is expressed as:

[0010] Wherein, i(t) represents the current value at time t, I0 represents the current peak value, k represents the correction coefficient, α represents the wavefront attenuation coefficient, β represents the wavetail attenuation coefficient, and t represents time.

[0011] Furthermore, the step of simulating the NEMP time domain standard waveform and the LEMP time domain standard waveform to obtain each simulated waveform includes: Taking time, position, equivalent, NEMP time domain standard waveform and LEMP time domain standard waveform as input parameters, and taking the NEMP waveform and LEMP waveform of the source area as reference waveforms at 0km, so as to use the reference waveforms to simulate the attenuation of the signal during propagation; Performing Fourier transformation on the NEMP waveform and the LEMP waveform of the source region to obtain the amplitude spectrum and the phase spectrum of the NEMP waveform and the LEMP waveform of the source region; The VLF signal earth-ionosphere channel propagation model is used to analyze the amplitude and phase spectra of the NEMP waveform and LEMP waveform in the source area, and the results of the amplitude and phase changes at different frequency points with distance are obtained. The results of the amplitude and phase changes with distance at different frequency points are superimposed with the amplitude spectrum and phase spectrum of the NEMP waveform and LEMP waveform in the source area to obtain the target amplitude and target phase of the NEMP waveform and LEMP waveform at the specified position, and generate the corresponding target amplitude spectrum and target phase spectrum; The target amplitude spectrum and the target phase spectrum are converted into target time domain waveforms through inverse Fourier transform to obtain various simulation waveforms.

[0012] Furthermore, in the step of normalizing the amplitude spectrum of each simulation waveform, the normalization is expressed as:

[0013] in, represents the amplitude spectrum, express The maximum value in represents the normalized result, i represents the i-th simulation waveform, f represents the frequency, and D represents the propagation distance.

[0014] Furthermore, in the step of determining a normalized total spectrum according to the distribution frequency band range of NEMP and the normalization processing result, the normalized total spectrum is expressed as:

[0015] in, Represents the normalized total spectrum. The distribution band of NEMP is 3kHz~60kHz. Represents the normalized result.

[0016] Furthermore, in the step of determining the characteristic value of each simulated waveform according to the normalized total spectrum, the spectrum center frequency is expressed as:

[0017] Indicates the frequency of the spectrum center; The half-peak width is expressed as:

[0018] represents the half peak width, for The inverse function of The mth maximum value in ; The average slope of the normalized spectrum front is expressed as:

[0019] represents the average slope of the normalized spectrum front, express The first maximum value greater than 0.5 from 3kHz to 60kHz; The average slope of the trailing edge of the normalized spectrum is expressed as:

[0020] represents the average slope of the trailing edge of the normalized spectrum, express The second maximum value greater than 0.5 from 3kHz to 60kHz; The normalized spectral skew is expressed as:

[0021] L represents the normalized spectral skew; The normalized spectral kurtosis is expressed as:

[0022] V represents the normalized spectral kurtosis.

[0023] A second aspect of an embodiment of the present invention provides an electromagnetic pulse identification system based on a naive Bayesian classification model, which is used to implement the electromagnetic pulse identification method based on a naive Bayesian classification model provided in the first aspect, and the system includes: A simulation module, used for simulating NEMP waveforms and LEMP waveforms to obtain various simulation waveforms, wherein the simulation waveforms include NEMP simulation waveforms and LEMP simulation waveforms; A normalization processing module is used to normalize the amplitude spectrum of each simulation waveform, and determine the normalized total spectrum according to the distribution frequency band range of NEMP and the normalization processing result; The characteristic value determination module is used to determine the characteristic values ​​of each simulation waveform according to the normalized total spectrum, wherein the characteristic values ​​include the spectrum center frequency, the half-peak width, the average slope of the normalized spectrum front edge, the average slope of the normalized spectrum rear edge, the normalized spectrum skew and the normalized spectrum kurtosis; An acquisition module is used to acquire a preset number of NEMP simulation waveforms and LEMP simulation waveforms, and determine the range of the target characteristic value according to the characteristic value distribution characteristics; A probability statistics module is used to calculate the probability of the target characteristic value taking [a+(i-1)×dx, a+i×dx), i=1, 2, ..., n for the characteristic values ​​of a preset number of NEMP simulation waveforms and LEMP simulation waveforms according to a preset step size, to obtain a first probability value representing NEMP and a second probability value representing LEMP, wherein a is the starting value of the target characteristic value range, and dx represents the preset step size; An identification module is used to identify an electromagnetic pulse according to the magnitude 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 a NEMP; when the first probability value is less than the second probability value, the electromagnetic pulse is a LEMP.

[0024] A third aspect of an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the electromagnetic pulse identification method based on the naive Bayes classification model provided in the first aspect.

[0025] The fourth aspect of an embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the electromagnetic pulse identification method based on the naive Bayes classification model provided in the first aspect is implemented.

[0026] An electromagnetic pulse identification method and system based on a naive Bayesian classification model is provided in an embodiment of the present invention. The method simulates a NEMP time domain standard waveform and a LEMP time domain standard waveform to obtain each simulation waveform; normalizes the amplitude spectrum of each simulation waveform, and determines the normalized total spectrum according to the distribution frequency band range of NEMP and the normalization processing result; determines the characteristic value of each simulation waveform according to the normalized total spectrum; obtains the characteristic values ​​of a preset number of NEMP simulation waveforms and LEMP simulation waveforms, selects the spectrum center frequency as the target characteristic value, and determines the range of the target characteristic value according to the electromagnetic pulse signal frequency range; according to the preset step size, respectively, for the characteristic values ​​of the preset number of NEMP simulation waveforms and LEMP simulation waveforms, the probability of the target characteristic value taking [a+(i-1)×dx, a+i×dx), i=1, 2,..., n, to obtain a first probability value representing NEMP and a second probability value representing LEMP, wherein a is the starting value of the target characteristic value range, and dx represents the preset step size; and identifies the electromagnetic pulse according to the size of the first probability value and the second probability value. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A flowchart of an implementation method of an electromagnetic pulse identification method based on a naive Bayesian classification model provided in the first embodiment of the present invention; Figure 2 A structural block diagram of an electromagnetic pulse identification system based on a naive Bayesian classification model provided in Embodiment 2 of the present invention; Figure 3 This is a structural block diagram of an electronic device provided in Embodiment 3 of the present invention. DETAILED DESCRIPTION

[0028] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0029] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0030] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0031] Embodiment 1 According to an embodiment of the present invention, a method for electromagnetic pulse identification based on a naive Bayes classification model is provided. 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 a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0032] In the first embodiment, an electromagnetic pulse recognition method based on a naive Bayesian classification model is provided, which can be used in electronic devices, such as computers. Figure 1 , Figure 1 A flowchart of an implementation of an electromagnetic pulse identification method based on a naive Bayesian classification model provided in the first embodiment of the present invention is shown, which specifically includes steps S01 to S06.

[0033] Step S01, simulating NEMP waveforms and LEMP waveforms to obtain various simulation waveforms, wherein the simulation waveforms include NEMP simulation waveforms and LEMP simulation waveforms.

[0034] Specifically, the NEMP time domain standard waveform is expressed as:

[0035] in, represents the electric field strength at time t, in volts per meter, represents the peak field intensity, m1 represents the parameters of the pulse front edge, n1 represents the parameters of the pulse rear edge, k1 represents the correction coefficient, and t represents the time in seconds. In this embodiment, E0=5000v / m, m1=4×10 7 s -1 , n1=6×10 8 s -1 , k1=1.3; The LEMP time domain standard waveform is expressed as:

[0036] Wherein, i(t) represents the current value at time t, I0 represents the current peak value, k represents the correction coefficient, α represents the wavefront attenuation coefficient, β represents the wavetail attenuation coefficient, and t represents time. In this embodiment, α is 7.714×10 4 , β is 2.489×10 5 , k is 2.33.

[0037] It should be noted that in order to establish a sample library, that is, a database composed of simulated waveforms, firstly, time, position, equivalent, NEMP time domain standard waveform and LEMP time domain standard waveform are used as input parameters, and the NEMP waveform and LEMP waveform of the source area are used as reference waveforms at 0km, so as to use the reference waveforms to simulate the attenuation of the signal during the propagation process; Performing Fourier transformation on the NEMP waveform and the LEMP waveform of the source region to obtain the amplitude spectrum and the phase spectrum of the NEMP waveform and the LEMP waveform of the source region; The VLF signal ground-ionosphere channel propagation model is used to analyze the amplitude spectrum and phase spectrum of the NEMP waveform and LEMP waveform in the source area, and the results of the amplitude and phase changes with distance at different frequency points are obtained. It should be noted that the earth's surface and the low ionosphere have excellent reflection performance for VLF electromagnetic waves. The distance between the low ionosphere and the ground is similar to the wavelength of VLF electromagnetic waves. The VLF electromagnetic waves propagate in the space formed by the earth and the ionosphere just like propagating in a waveguide. The VLF ground-ionosphere channel propagation model is a model for studying the propagation characteristics of VLF electromagnetic waves in the waveguide structure formed between the earth's surface and the ionosphere. The results of the amplitude and phase changes with distance at different frequency points are superimposed with the amplitude spectrum and phase spectrum of the NEMP waveform and LEMP waveform in the source area to obtain the target amplitude and target phase of the NEMP waveform and LEMP waveform at the specified position, and generate the corresponding target amplitude spectrum and target phase spectrum; The target amplitude spectrum and the target phase spectrum are converted into target time domain waveforms through inverse Fourier transform to obtain various simulation waveforms.

[0038] In some other embodiments of the present invention, in order to make the simulation waveform cover different conditions as much as possible, the source points and receiving points can be selected evenly in space, the date selection can cover different seasons, and the local time can cover different typical times of the day and night. This is done to ensure that the simulation samples have good temporal and spatial coverage.

[0039] Step S02, normalizing the amplitude spectra of each simulation waveform, and determining a normalized total spectrum according to the distribution frequency band range of NEMP and the normalization processing result.

[0040] Specifically, if E i (t, D) represents the waveform of the i-th sample, and its amplitude spectrum can be expressed as (f,D), and its normalization is expressed as:

[0041] in, represents the amplitude spectrum, express The maximum value in represents the normalized result, i represents the i-th simulation waveform, f represents the frequency, and D represents the propagation distance.

[0042] Since the main components of the long-distance waveform of the nuclear explosion electromagnetic pulse are concentrated in the frequency band of 0kHz~60kHz, the following characteristic quantities are defined in this frequency band.

[0043] The normalized score is expressed as:

[0044] in, Represents the normalized total spectrum. The distribution band of NEMP is 3kHz~60kHz. Represents the normalized result.

[0045] Step S03, determining the characteristic values ​​of each simulation waveform according to the normalized total spectrum, wherein the characteristic values ​​include spectrum center frequency, half-peak width, normalized spectrum leading edge average slope, normalized spectrum trailing edge average slope, normalized spectrum skew and normalized spectrum kurtosis.

[0046] Specifically, the spectrum center frequency is expressed as:

[0047] Indicates the frequency of the spectrum center; The half-peak width is expressed as:

[0048] represents the half peak width, for The inverse function of In addition, the summation symbol is for When the maximum value in is not 1, it should be the sum of the half-peak widths corresponding to the extreme values. In particular, for the normalized spectrum, it can be written as:

[0049] Number of maximum values ​​greater than half the peak:

[0050] When n i When only 1 or 2 is taken, the average slope of the front and rear edges can be defined by taking the outside of the two maximum values. When n is greater than 2, this definition does not hold; The average slope of the normalized spectrum front is expressed as:

[0051] represents the average slope of the normalized spectrum front, express The first maximum value greater than 0.5 from 3kHz to 60kHz; The average slope of the trailing edge of the normalized spectrum is expressed as:

[0052] represents the average slope of the trailing edge of the normalized spectrum, express The second maximum value greater than 0.5 from 3kHz to 60kHz; The normalized spectral skew is expressed as:

[0053] L represents the normalized spectral skew; The normalized spectral kurtosis is expressed as:

[0054] V represents the normalized spectral kurtosis.

[0055] Step S04, obtaining a preset number of characteristic values ​​of NEMP simulation waveforms and LEMP simulation waveforms, and determining the range of the target characteristic value according to the characteristic value distribution characteristics.

[0056] Among them, characteristic values ​​of a preset number of NEMP simulation waveforms and LEMP simulation waveforms are obtained, the spectrum 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.

[0057] In this embodiment, 10,000,000 NEMP samples are obtained, recorded as ω1, and 10,000,000 LEMP samples are obtained, recorded as ω2, where the sample is the simulation waveform, and a normalized spectrum analysis is performed on each NEMP and LEMP sample, and the spectrum center frequency data is extracted as x. According to the frequency range of the electromagnetic pulse signal, the range of x is limited to 3kHz~60kHz.

[0058] Step S05, according to a preset step size, respectively, for a preset number of NEMP simulation waveforms and LEMP simulation waveforms, the probability of the target characteristic value taking [a+(i-1)×dx, a+i×dx), i=1, 2, ..., n is counted to obtain a first probability value representing NEMP and a second probability value representing LEMP, wherein a is the starting value of the target characteristic value range, and dx represents the preset step size.

[0059] Taking the spectral center frequency as an example, the frequency range is 3~60kHz, the step size dx can be set to 0.1kHz, and the spectral center frequency characteristic value statistical interval is [3+(i-1)×dx, 3+i×dx), i=1, 2,..., 570.

[0060] Step S06, identifying the electromagnetic pulse according to 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; when the first probability value is less than the second probability value, the electromagnetic pulse is LEMP.

[0061] It should be noted that, assuming there is an unknown sample, first extract its spectral center frequency x, and substitute it into p(x|ω1) and p(x|ω2) to obtain the likelihood ratio of the sample:

[0062] The likelihood ratio threshold can be calculated directly based on the number of samples in the two categories used to obtain the p(x|ω1) and p(x|ω2) curves:

[0063] Since both NEMP and LEMP samples are 10,000,000, At this time, the threshold is the intersection of the two curves. By directly comparing the sizes of p(x|ω1) and p(x|ω2), the minimum error rate Bayesian decision can be completed. The larger one is the one that is judged as that category.

[0064] In order to evaluate the performance of the above decision-making method, we first calculate the first type error rate and the second type error rate and draw the ROC curve. Specifically, the first type error rate is the false positive rate, that is, the probability of a non-nuclear explosion sample being mistakenly reported as a nuclear explosion event, also known as the false alarm rate or false alarm rate, expressed as:

[0065] That is, the error rate of deciding the second category samples as the first category.

[0066] The second type of error rate is the false negative rate, that is, the probability of a nuclear explosion sample being mistakenly reported as a non-nuclear explosion event, also known as the missed alarm rate, expressed as:

[0067] That is, the error rate of deciding the first type of samples as the second type.

[0068] Since the threshold is determined, the boundaries of R1 and R2 are also determined. 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:

[0069] In actual calculation, since p(x|ω1) and p(x|ω2) are discrete values, we can simply sum them with the cutoff point as the boundary. Note that the ordinate axis of the ROC curve is the sensitivity Sn, that is, the true positive rate, Sn=1-P1(e); the abscissa axis is the false positive rate, which is P2(e).

[0070] By controlling the prior probability, that is, the number of samples in the two categories, the threshold size can be adjusted. In one extreme case, all samples are negative samples, that is, LEMP samples, at which p(x|ω1)=0. At this time, any unknown sample will be judged as a negative sample, the false positive rate is 0, and the true positive rate is also 0, and the (0, 0) point is obtained on the ROC curve. In the other extreme case, all samples are positive samples, that is, NEMP samples, at which p(x|ω2)=0. At this time, any unknown sample will be judged as a positive sample, the true positive rate is 1, and the false positive rate is also 1, and the (1, 1) point is obtained on the ROC curve.

[0071] N1 and N2 represent the number of samples of NEMP and LEMP respectively. The total number of samples is controlled at 1 million. The ratio of N1 and N2 is gradually adjusted to change the threshold value. For example, with a step size of 10,000, the ratios are set to 0:100, 1:99, 2:98, 3:97, ..., 99:1, 100:0, in units of 10,000. The above settings will form a total of 101 prior probabilities, corresponding to 101 thresholds. Each threshold can obtain two types of error rates, and then the ROC curve can be drawn.

[0072] Furthermore, for the classifier composed of a single feature parameter, let the probability of correct identification of NEMP by the minimum error rate Bayesian decision be P (i.e., the recognition rate be P). If n stations simultaneously collect the event, there will be n waveforms. Since they are the same parameters, their monitoring performance is the same. The theoretical recognition accuracy of nuclear explosion electromagnetic pulses through voting fusion is: P=P n +P n-1 (1-P)×n+...+P n / 2 (1-P) n-n / 2 ×n Calculation example: Assuming that there are 5 monitoring stations, the recognition rates of the 4 characteristic parameters are 99%, 98%, 95%, and 90% respectively. The recognition rate obtained by multi-parameter voting fusion is: The probability of correctly identifying all four feature parameters: 99%×98%×95%×90%=82.9521%; The probability of correct identification of the three characteristic parameters: 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%; The probability of correct identification of 2 feature parameters: 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%; The comprehensive recognition rate is: 99.0657%+0.6844%=99.7501%.

[0073] In summary, the electromagnetic pulse identification method based on the naive Bayes classification model in the above-mentioned embodiment of the present invention simulates the NEMP time domain standard waveform and the LEMP time domain standard waveform to obtain each simulation waveform; normalizes the amplitude spectrum of each simulation waveform, and determines the normalized total spectrum according to the distribution frequency band range of NEMP and the normalization processing result; determines the characteristic value of each simulation waveform according to the normalized total spectrum; obtains the characteristic values ​​of a preset number of simulation waveforms, selects the spectrum center frequency as the target characteristic value, and determines the range of the target characteristic value according to the frequency range of the electromagnetic pulse signal; according to the preset step size, respectively counts the probability when the target characteristic value takes the preset value for the characteristic values ​​of the preset number of simulation waveforms, and obtains a first probability value representing NEMP and a second probability value representing LEMP; identifies the electromagnetic pulse according to the size of the first probability value and the second probability value.

[0074] Embodiment 2 See also Figure 2 , Figure 2 This is a structural block diagram of an electromagnetic pulse identification system based on a naive Bayesian classification model provided in the second embodiment of the present invention. The electromagnetic pulse identification system 200 based on the naive Bayesian classification model is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0075] Specifically, the electromagnetic pulse identification system 200 based on the naive Bayes classification model includes: a simulation module 21, a normalization processing module 22, a feature value determination module 23, an acquisition module 24, a probability statistics module 25 and an identification module 26, wherein: The simulation module 21 is used to simulate the NEMP waveform and the LEMP waveform to obtain each simulation waveform, wherein the simulation waveform includes the NEMP simulation waveform and the LEMP simulation waveform, and the NEMP time domain standard waveform is expressed as:

[0076] in, represents the electric field strength at time t, represents the peak field intensity, m1 represents the parameters of the pulse front edge, n1 represents the parameters of the pulse trailing edge, k1 represents the correction coefficient, t represents the time, and the LEMP time domain standard waveform is expressed as:

[0077] Where i(t) represents the current value at time t, I0 represents the current peak value, k represents the correction coefficient, α represents the wavefront attenuation coefficient, β represents the wavetail attenuation coefficient, and t represents time; The normalization processing module 22 is used to normalize the amplitude spectrum of each simulation waveform, and determine the normalized total spectrum according to the distribution frequency band range of NEMP and the normalization processing result. The normalization processing is expressed as:

[0078] in, represents the amplitude spectrum, express The maximum value in represents the normalized result, i represents the i-th simulation waveform, f represents the frequency, D represents the propagation distance, and the normalized total spectrum is expressed as:

[0079] in, Represents the normalized total spectrum. The distribution band of NEMP is 3kHz~60kHz. represents the normalized result; The characteristic value determination module 23 is used to determine the characteristic values ​​of each simulation waveform according to the normalized total spectrum. The characteristic values ​​include the spectrum center frequency, half-peak width, the average slope of the normalized spectrum front edge, the average slope of the normalized spectrum rear edge, the normalized spectrum skew and the normalized spectrum kurtosis. The spectrum center frequency is expressed as:

[0080] Indicates the frequency of the spectrum center; The half-peak width is expressed as:

[0081] represents the half peak width, for The inverse function of The mth maximum value in ; The average slope of the normalized spectrum front is expressed as:

[0082] represents the average slope of the normalized spectrum front, express The first maximum value greater than 0.5 from 3kHz to 60kHz; The average slope of the trailing edge of the normalized spectrum is expressed as:

[0083] represents the average slope of the trailing edge of the normalized spectrum, express The second maximum value greater than 0.5 from 3kHz to 60kHz; The normalized spectral skew is expressed as:

[0084] L represents the normalized spectral skew; The normalized spectral kurtosis is expressed as:

[0085] V represents the normalized spectral kurtosis; An acquisition module 24 is used to acquire a preset number of characteristic values ​​of NEMP simulation waveforms and LEMP simulation waveforms, and determine the range of the target characteristic value according to the characteristic value distribution characteristics; The probability statistics module 25 is used to calculate the probability of the target characteristic value taking [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 the preset step length, to obtain a first probability value representing NEMP and a second probability value representing LEMP, wherein a is the starting value of the target characteristic value range, and dx represents the preset step length; The identification module 26 is used to identify the electromagnetic pulse according to the size 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; when the first probability value is less than the second probability value, the electromagnetic pulse is LEMP.

[0086] Further, in some optional embodiments of the present invention, the simulation module 21 includes: A definition unit is used to take time, position, equivalent, NEMP time domain standard waveform and LEMP time domain standard waveform as input parameters, and take the NEMP waveform and LEMP waveform of the source area as reference waveforms at 0km, so as to use the reference waveforms to simulate the attenuation of the signal during the propagation process; A Fourier transform unit, used for performing Fourier transform on the NEMP waveform and the LEMP waveform of the source region to obtain the amplitude spectrum and the phase spectrum of the NEMP waveform and the LEMP waveform of the source region; An analysis unit is used to analyze the amplitude spectrum and phase spectrum of the NEMP waveform and the LEMP waveform in the source area by adopting a very low frequency signal earth-ionosphere channel propagation model, and obtain the results of the amplitude and phase changes of different frequency points with distance; A superposition unit is used to superimpose the results of the amplitude and phase changes of different frequency points with the distance with the amplitude spectrum and phase spectrum of the NEMP waveform and the LEMP waveform of the source area, obtain the target amplitude and target phase of the NEMP waveform and the LEMP waveform at the specified position, and generate the corresponding target amplitude spectrum and target phase spectrum; The inverse Fourier transform unit is used to transform the target amplitude spectrum and the target phase spectrum into a target time domain waveform through inverse Fourier transform to obtain each simulation waveform.

[0087] Embodiment 3 Another aspect of the present invention provides an electronic device, see Figure 3 , shown is an electronic device in Embodiment 3 of the present invention, including a memory 20, a processor 10, and a computer program 30 stored in the memory and executable on the processor. When the processor 10 executes the computer program 30, the electromagnetic pulse identification method based on the naive Bayes classification model as described above is implemented.

[0088] In some embodiments, the processor 10 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor or other data processing chip, used to run program codes or process data stored in the memory 20, such as executing access restriction programs.

[0089] The memory 20 includes at least one type of readable storage medium, and the readable storage medium includes a flash memory, a hard disk, a multimedia card, a card-type memory (for example, an SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory 20 may be an internal storage unit of an electronic device, such as a hard disk of the electronic device. In other embodiments, the memory 20 may also be an external storage device of an electronic device, such as a plug-in hard disk equipped on the electronic device, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. Further, the memory 20 may also include both an internal storage unit and an external storage device of the electronic device. The memory 20 may be used not only to store application software and various types of data of the electronic device, but also to temporarily store data that has been output or is to be output.

[0090] It should be pointed out that Figure 3 The structure shown does not constitute a limitation on the electronic device. In other embodiments, the electronic device may include fewer or more components than those shown in the figure, or combine certain components, or arrange the components differently.

[0091] The embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the electromagnetic pulse identification method based on the naive Bayes classification model as described above is implemented.

[0092] Those skilled in the art will appreciate that the logic and / or steps represented in the flowchart or otherwise described herein, for example, may be considered as an ordered list of executable instructions for implementing logical functions, and may be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For purposes of this specification, "computer-readable medium" may be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0093] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0094] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or a combination thereof: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0095] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0096] The above embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present invention. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention patent shall be subject to the attached claims.

Claims

1. An electromagnetic pulse recognition method based on a naive Bayesian classification model, characterized in that: The method comprises: Conducting simulations of NEMP waveforms and LEMP waveforms to obtain various simulation waveforms, wherein the simulation waveforms include NEMP simulation waveforms and LEMP simulation waveforms; The amplitude spectrum of each simulation waveform is normalized, and the normalized total spectrum is determined according to the distribution frequency band range of NEMP and the normalization processing result; According to the normalized total spectrum, the characteristic values ​​of each simulation waveform are determined, wherein the characteristic values ​​include the spectrum center frequency, the half-peak width, the average slope of the normalized spectrum front edge, the average slope of the normalized spectrum rear edge, the normalized spectrum skew and the normalized spectrum kurtosis; Acquire a preset number of NEMP simulation waveforms and LEMP simulation waveforms, and determine the range of the target characteristic value according to the characteristic value distribution characteristics; According to a preset step size, the probability of the target characteristic value taking [a+(i-1)×dx, a+i×dx), i=1, 2, ..., n, is counted for the characteristic values ​​of a preset number of NEMP simulation waveforms and LEMP simulation waveforms respectively, to obtain a first probability value representing NEMP and a second probability value representing LEMP, wherein a is the starting value of the target characteristic value range, and dx represents the preset step size; The electromagnetic pulse is identified according to the magnitude 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; when the first probability value is less than the second probability value, the electromagnetic pulse is LEMP.

2. The electromagnetic pulse identification method based on the naive Bayes classification model according to claim 1 is characterized in that: The NEMP time domain standard waveform is expressed as: in, represents the electric field strength at time t, represents the peak field strength, m1 represents the parameters of the pulse leading edge, n1 represents the parameters of the pulse trailing edge, k1 represents the correction coefficient, and t represents time.

3. The electromagnetic pulse identification method based on the naive Bayes classification model according to claim 2 is characterized in that: The LEMP time domain standard waveform is expressed as: Wherein, i(t) represents the current value at time t, I0 represents the current peak value, k represents the correction coefficient, α represents the wavefront attenuation coefficient, β represents the wavetail attenuation coefficient, and t represents time.

4. The electromagnetic pulse identification method based on the naive Bayes classification model according to claim 3 is characterized in that: The steps of simulating the NEMP waveform and the LEMP waveform to obtain each simulated waveform include: Taking time, position, equivalent, NEMP time domain standard waveform and LEMP time domain standard waveform as input parameters, and taking the NEMP waveform and LEMP waveform of the source area as reference waveforms at 0km, so as to use the reference waveforms to simulate the attenuation of the signal during propagation; Performing Fourier transformation on the NEMP waveform and the LEMP waveform of the source region to obtain the amplitude spectrum and the phase spectrum of the NEMP waveform and the LEMP waveform of the source region; The VLF signal earth-ionosphere channel propagation model is used to analyze the amplitude and phase spectra of the NEMP waveform and LEMP waveform in the source area, and the results of the amplitude and phase changes at different frequency points with distance are obtained. The results of the amplitude and phase changes with distance at different frequency points are superimposed with the amplitude spectrum and phase spectrum of the NEMP waveform and LEMP waveform in the source area to obtain the target amplitude and target phase of the NEMP waveform and LEMP waveform at the specified position, and generate the corresponding target amplitude spectrum and target phase spectrum; The target amplitude spectrum and the target phase spectrum are converted into target time domain waveforms through inverse Fourier transform to obtain various simulation waveforms.

5. The electromagnetic pulse identification method based on the naive Bayes classification model according to claim 4 is characterized in that: In the step of normalizing the amplitude spectrum of each simulation waveform, the normalization is expressed as: in, represents the amplitude spectrum, express The maximum value in represents the normalized result, i represents the i-th simulation waveform, f represents the frequency, and D represents the propagation distance.

6. The electromagnetic pulse identification method based on the naive Bayes classification model according to claim 5 is characterized in that: In the step of determining the normalized total spectrum according to the distribution frequency band range of NEMP and the normalization processing result, the normalized total spectrum is expressed as: in, Represents the normalized total spectrum. The distribution band of NEMP is 3kHz~60kHz. Represents the normalized result.

7. The electromagnetic pulse identification method based on the naive Bayes classification model according to claim 6 is characterized in that: In the step of determining the characteristic value of each simulated waveform according to the normalized total spectrum, the spectrum center frequency is expressed as: Indicates the frequency of the spectrum center; The half-peak width is expressed as: represents the half peak width, for The inverse function of The mth maximum value in ; The average slope of the normalized spectrum front is expressed as: represents the average slope of the normalized spectrum front, express The first maximum value greater than 0.5 from 3kHz to 60kHz; The average slope of the trailing edge of the normalized spectrum is expressed as: represents the average slope of the trailing edge of the normalized spectrum, express The second maximum value greater than 0.5 from 3kHz to 60kHz; The normalized spectral skew is expressed as: L represents the normalized spectral skew; The normalized spectral kurtosis is expressed as: V represents the normalized spectral kurtosis.

8. An electromagnetic pulse identification system based on a naive Bayesian classification model, characterized in that: The system is used to implement the electromagnetic pulse identification method based on the naive Bayes classification model as described in any one of claims 1 to 7, the system comprising: A simulation module, used for simulating NEMP waveforms and LEMP waveforms to obtain various simulation waveforms, wherein the simulation waveforms include NEMP simulation waveforms and LEMP simulation waveforms; A normalization processing module is used to normalize the amplitude spectrum of each simulation waveform, and determine the normalized total spectrum according to the distribution frequency band range of NEMP and the normalization processing result; The characteristic value determination module is used to determine the characteristic values ​​of each simulation waveform according to the normalized total spectrum, wherein the characteristic values ​​include the spectrum center frequency, the half-peak width, the average slope of the normalized spectrum front edge, the average slope of the normalized spectrum rear edge, the normalized spectrum skew and the normalized spectrum kurtosis; An acquisition module is used to acquire a preset number of NEMP simulation waveforms and LEMP simulation waveforms, and determine the range of the target characteristic value according to the characteristic value distribution characteristics; A probability statistics module is used to calculate the probability of the target characteristic value taking [a+(i-1)×dx, a+i×dx), i=1, 2, ..., n for the characteristic values ​​of a preset number of NEMP simulation waveforms and LEMP simulation waveforms according to a preset step size, to obtain a first probability value representing NEMP and a second probability value representing LEMP, wherein a is the starting value of the target characteristic value range, and dx represents the preset step size; An identification module is used to identify an electromagnetic pulse according to the magnitude 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 a NEMP; when the first probability value is less than the second probability value, the electromagnetic pulse is a LEMP.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the electromagnetic pulse identification method based on the naive Bayes classification model as described in any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the program, the electromagnetic pulse identification method based on the naive Bayes classification model as described in any one of claims 1 to 7 is implemented.

Images