A method for identifying fast radio bursts and cellular radio frequency interference

Abstract

This application relates to the field of radio astronomy and discloses a method for identifying fast radio bursts (FRBs) and cellular radio frequency interference (RFI). The method includes: acquiring multi-subband signals and constructing a signal matrix and a broadband signal based on the multi-subband signals; generating a dispersion curve matching degree based on the estimated and theoretical time delay vectors of the signal matrix; generating a first decision value based on the dispersion curve matching degree; generating a second decision value based on the mean of the frequency domain narrowband entropy; generating a third decision value based on the time domain broadband entropy and the short-time Fourier transform entropy; generating a fourth decision value based on the mean of the Welch power spectrum narrowband entropy and the Welch power spectrum broadband entropy; generating a fifth decision value based on the frequency domain broadband entropy; and identifying the FRB or cellular RPI based on the weighted summation result. Its beneficial effect is a reduction in the FRB misclassification rate.

Description

Technical Field

[0001] This application relates to the field of radio astronomy technology, and in particular to a method for identifying fast radio bursts and cellular radio frequency interference. Background Technology

[0002] With the widespread adoption of 5G and 6G communication technologies, cellular radio frequency interference has become an increasingly prominent issue affecting the detection of Fast Radio Bursts (FRBs). The transient characteristics and time-domain dispersion range of these interference signals highly overlap with those of FRBs, making them a major source of false alarms in FRB detection. Current techniques relying solely on dispersion curve analysis cannot effectively distinguish between genuine dispersion from cosmological sources and the pseudo-dispersion characteristics of man-made interference signals. Furthermore, traditional entropy feature analysis uses only a single entropy index (such as Shannon entropy), which has weak resolution capabilities for overlapping pulse signals and lacks integration with dispersion information to form a comprehensive judgment logic. Therefore, current techniques have significant limitations in accurately identifying FRBs and cellular radio frequency interference. Summary of the Invention

[0003] This application provides a method for identifying fast radio bursts and cellular radio frequency interference, which solves the technical problem of how to accurately identify fast radio bursts and cellular radio frequency interference, and achieves the technical effect of reducing the false positive rate.

[0004] To achieve the above objectives, the main technical solutions adopted in this application include:

[0005] In a first aspect, embodiments of this application provide a method for identifying fast radio bursts and cellular radio frequency interference. The method includes: acquiring multiple sub-band signals and constructing a signal matrix and a broadband signal based on the multiple sub-band signals, wherein the columns of the signal matrix correspond to sub-bands with different center frequencies, and the broadband signal is generated by superimposing the multiple sub-band signals; generating a dispersion curve matching degree based on the estimated time delay vector and the theoretical time delay vector of the signal matrix; extracting the frequency domain narrowband entropy and the Welch power spectrum narrowband entropy for any sub-band in the signal matrix; determining the average frequency domain narrowband entropy based on the frequency domain narrowband entropy of each sub-band; determining the average Welch power spectrum narrowband entropy based on the Welch power spectrum narrowband entropy of each sub-band; and for the broadband signal... With the signal, the time-domain broadband entropy, frequency-domain broadband entropy, short-time Fourier transform entropy, and Welch power spectrum broadband entropy are extracted respectively. Based on the dispersion curve matching degree, a first decision value is generated; based on the mean of the frequency-domain narrowband entropy, a second decision value is generated; based on the time-domain broadband entropy and the short-time Fourier transform entropy, a third decision value is generated; based on the mean of the Welch power spectrum narrowband entropy and the Welch power spectrum broadband entropy, a fourth decision value is generated; based on the frequency-domain broadband entropy, a fifth decision value is generated; the first, second, third, fourth, and fifth decision values ​​are weighted and summed, and the fast radio burst or cellular radio frequency interference is identified based on the weighted summation result.

[0006] The method for identifying Fast Radio Bursts (FRBs) and Cellular Radio Frequency Interference (RFI) proposed in this application integrates dispersion curve matching degree and multi-entropy features, including frequency domain narrowband entropy, Welch power spectrum narrowband entropy, time domain broadband entropy, frequency domain broadband entropy, short-time Fourier transform entropy, and Welch power spectrum broadband entropy, to comprehensively capture the essential differences between FRBs and cellular RFIs. It utilizes multi-layer logic decision-making to distinguish between FRBs and cellular RFIs, solving the technical problem of how to accurately identify FRBs and cellular RFIs and achieving the technical effect of reducing the false positive rate.

[0007] Optionally, the theoretical time delay vector is determined as follows: for any candidate dispersion value, the theoretical time delay of each sub-band in the signal matrix is ​​determined; for any sub-band, time delay compensation is performed based on the theoretical time delay to obtain the compensated sub-band signal; the compensated sub-band signals are merged to obtain a merged signal, and the signal-to-noise ratio (SNR) of the merged signal is calculated; the maximum SNR among the various SNRs is identified, and the candidate dispersion value corresponding to the maximum SNR is taken as the target dispersion value; based on the target dispersion value, the target theoretical time delay of each sub-band is determined; based on the target theoretical time delay of each sub-band, the theoretical time delay vector of the signal matrix is ​​determined.

[0008] Optionally, the estimated time delay vector is determined as follows: for any sub-band, the peak position index is corrected using the three-point parabolic interpolation method to obtain the corrected index; the estimated time delay of the sub-band is calculated based on the corrected index; and the estimated time delay vector of the signal matrix is ​​generated based on the estimated time delay of each sub-band.

[0009] Optionally, the method further includes: for any sub-band, converting the theoretical time delay into the number of sampling points; performing time delay compensation on the sub-band based on the number of sampling points to obtain a compensated sub-band signal, wherein the length of each compensated sub-band signal is the same.

[0010] Optionally, the method further includes: if the dispersion curve matching degree is greater than or equal to a first threshold, then the first decision value is set to 1; if the mean value of the frequency domain narrowband entropy is greater than or equal to a second threshold, then the second decision value is set to 1; if the time domain broadband entropy is greater than a third threshold and the short-time Fourier transform entropy is greater than a fourth threshold, then the third decision value is set to 1; if the mean value of the Welch power spectrum narrowband entropy is greater than or equal to a fifth threshold or the Welch power spectrum broadband entropy is greater than or equal to a sixth threshold, then the fourth decision value is set to 1; if the frequency domain broadband entropy is greater than a seventh threshold, then the fifth decision value is set to 1.

[0011] Optionally, the first threshold is 0.8, the second threshold is 2, the third threshold is 6, the fourth threshold is 5.5, the fifth threshold is 3, the sixth threshold is 4, and the seventh threshold is 5.

[0012] Optionally, a weighted summation is performed on the first decision value, the second decision value, the third decision value, the fourth decision value, and the fifth decision value, including: setting the weight of the first decision value as a first weight, setting the weight of the second decision value as a second weight, and setting the weights of the third decision value, the fourth decision value, and the fifth decision value as a third weight, wherein the first weight is greater than the second weight, and the second weight is greater than the third weight.

[0013] Optionally, the first weight is 0.5, the second weight is 0.2, and the third weight is 0.1.

[0014] Optionally, if the weighted summation result is greater than or equal to a first preset value, it is determined to be a fast radio burst; if the weighted summation result is less than or equal to a second preset value, it is determined to be cellular radio frequency interference.

[0015] Optionally, the first setting value is 0.7, and the second setting value is 0.3.

[0016] Secondly, embodiments of this application provide a system for identifying fast radio bursts and cellular radio frequency interference. The system includes: an acquisition module for acquiring multiple sub-band signals and constructing a signal matrix and a broadband signal based on the multiple sub-band signals, wherein the columns of the signal matrix correspond to sub-bands with different center frequencies, and the broadband signal is generated by superimposing the multiple sub-band signals; a generation module for generating a dispersion curve matching degree based on the estimated time delay vector and the theoretical time delay vector of the signal matrix; and an extraction module for extracting the frequency domain narrowband entropy and Welch power spectrum narrowband entropy for any sub-band in the signal matrix; for determining the average frequency domain narrowband entropy based on the frequency domain narrowband entropy of each sub-band; for determining the average Welch power spectrum narrowband entropy based on the Welch power spectrum narrowband entropy of each sub-band; and for... The broadband signal is used to extract time-domain broadband entropy, frequency-domain broadband entropy, short-time Fourier transform entropy, and Welch power spectrum broadband entropy, respectively. A decision module is used to generate a first decision value based on the dispersion curve matching degree; a second decision value based on the mean frequency-domain narrowband entropy; a third decision value based on the time-domain broadband entropy and the short-time Fourier transform entropy; a fourth decision value based on the mean Welch power spectrum narrowband entropy and the Welch power spectrum broadband entropy; and a fifth decision value based on the frequency-domain broadband entropy. An identification module is used to perform a weighted summation of the first, second, third, fourth, and fifth decision values, and to identify fast radio bursts or cellular radio frequency interference based on the weighted summation result.

[0017] Thirdly, embodiments of this application provide a computer device, including: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the above-mentioned method for identifying fast radio bursts and cellular radio frequency interference.

[0018] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer instructions, which are used to cause a computer to execute the above-described method for identifying fast radio bursts and cellular radio frequency interference.

[0019] Fifthly, embodiments of this application provide a computer program product, including computer instructions, which are used to cause a computer to execute the above-described method for identifying fast radio bursts and cellular radio frequency interference. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0021] Figure 1 A flowchart illustrating the method for identifying fast radio bursts and cellular radio frequency interference provided in this application embodiment;

[0022] Figure 2 A schematic diagram of a fast radio burst and cellular radio frequency interference identification system provided in an embodiment of this application;

[0023] Figure 3 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0025] Fast radio bursts (FRBs) are high-energy radio signals originating from deep space, with the entire burst lasting only a few milliseconds. These signals have extremely high dispersion, typically exceeding 100. Since their first discovery in 2007, FRBs have become a focus of astronomical research. Detecting FRBs presents researchers with numerous challenges. First, although the energy emitted from their sources is extremely high, the signal is extremely weak by the time it reaches the receiver on Earth. Second, except for a few extremely short cases with errors, the duration of most discovered FRBs is typically on the order of milliseconds, with typical durations of repeating FRBs ranging from 0.1 to 10 milliseconds and non-repeating FRBs typically ranging from 1 to 5 milliseconds. Third, due to scattering by the interstellar medium, the received signal exhibits severe broadening. Fourth, traditional detection methods relying on dispersion measurements (DM) are susceptible to man-made broadband interference in the low-frequency radio band (300 MHz–1.5 GHz), leading to a higher false alarm rate.

[0026] With the widespread application of 5G and 6G communication technologies, cellular frequency interference (RFI) has increasingly impacted the detection of fast radio bursts (FRBs). 5G NR signals employ burst transmission modes, generating millisecond-level transient pulses whose time-frequency characteristics are remarkably similar to those of FRBs. Especially in low-Earth orbit satellite communication scenarios, the dispersion measure (DM value) of 5G signals can reach 200–500, overlapping with the dispersion range of real FRBs. Furthermore, the significantly improved sensitivity of next-generation radio telescopes has made previously negligible cellular interference, such as base station sidelobe radiation, a major source of false alarms. According to statistics from the Parkes telescope, approximately 35% of FRB candidate signals are ultimately confirmed to be ground-based interference.

[0027] Currently, there are significant shortcomings in the technical means to distinguish between fast radio bursts and interference signals. Related technologies rely solely on analyzing dispersion curves, which cannot differentiate between cosmological dispersion and man-made signals. Traditional entropy feature analysis uses only a single entropy index (such as Shannon entropy), which has limited ability to distinguish overlapping pulse signals and does not combine dispersion information for comprehensive judgment, leading to easy misjudgment.

[0028] The method for identifying fast radio bursts (FRBs) and cellular radio frequency interference (RFI) proposed in this application integrates multiple features such as dispersion measurement, time-domain energy entropy, frequency-domain entropy, and time-frequency domain entropy to construct a multi-dimensional discrimination system. This system comprehensively captures the essential differences between FRBs and cellular RFIs and uses multi-layer logic decision-making to distinguish between FRBs and cellular RFIs. This solves the technical problem of how to accurately identify FRBs and cellular RFIs and achieves the technical effect of reducing the false positive rate.

[0029] This application provides a method for identifying fast radio bursts and cellular radio frequency interference. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0030] Please refer to Figure 1 , Figure 1 A flowchart illustrating the method for identifying fast radio bursts and cellular radio frequency interference provided in this application embodiment is shown below. Figure 1 As shown, the process includes the following steps:

[0031] Step S100: Obtain multiple sub-band signals and construct a signal matrix and a broadband signal based on the multiple sub-band signals. The columns of the signal matrix correspond to sub-bands with different center frequencies, and the broadband signal is generated by superimposing the multiple sub-band signals.

[0032] Among them, based on The signal matrix constructed by the path signal can be represented as:

[0033]

[0034] In the above formula, for matrix, The number of sub-band signals. This refers to the length of a single signal channel. Indicates the first The sub-band signal has a bandwidth of [missing information]. , No. The center frequency of the band is: ,and It is the center frequency of the 0th subband signal.

[0035] in, , Representing vectors The nth element.

[0036] Will The signal is generated by superimposing the signal from the path. , can be represented as:

[0037]

[0038] In the above formula, .

[0039] Step S300: Based on the estimated time delay vector and the theoretical time delay vector of the signal matrix, the dispersion curve matching degree is obtained.

[0040] The elements of the estimated delay vector are composed of the estimated delays of each sub-band in the signal matrix. The elements of the theoretical delay vector are composed of the target theoretical delays of each sub-band in the signal matrix. For any sub-band, the theoretical delay refers to the delay of that sub-band relative to the 0th sub-band signal, which is the sub-band with the highest frequency.

[0041] The correlation coefficient between the estimated time delay vector and the theoretical time delay vector is calculated, which yields the dispersion curve matching degree. FRBs have high dispersion characteristics, and the dispersion curve matching degree is the core physical basis for identifying FRBs. However, the dispersion curve of cellular RFIs usually does not match the theoretical cosmological model.

[0042] Step S500: For any sub-band in the signal matrix, extract the frequency domain narrowband entropy and the Welch power spectrum narrowband entropy respectively; determine the mean value of the frequency domain narrowband entropy based on the frequency domain narrowband entropy of each sub-band; determine the mean value of the Welch power spectrum narrowband entropy based on the Welch power spectrum narrowband entropy of each sub-band; for the broadband signal, extract the time domain broadband entropy, frequency domain broadband entropy, short-time Fourier transform entropy, and Welch power spectrum broadband entropy respectively.

[0043] The frequency domain narrowband entropy is the entropy of each sub-band calculated through its frequency domain energy distribution, used to characterize the overall energy distribution uniformity of the sub-band signal in the instantaneous frequency domain. The mean frequency domain narrowband entropy is the average of the frequency domain narrowband entropies of all sub-bands.

[0044] Welch power spectrum narrowband entropy is used to characterize the uniformity of the frequency domain power distribution of a sub-band. A higher entropy value indicates a more uniform power distribution across different frequencies. The mean Welch power spectrum narrowband entropy is calculated by summing the Welch power spectrum narrowband entropies of each sub-band and dividing by the total number of sub-bands. Since FRB signals and cellular RFI signals differ in their frequency domain power distribution characteristics, analyzing the mean Welch power spectrum narrowband entropy helps determine whether a signal belongs to FRB or cellular RFI. A higher mean indicates a more uniform overall frequency domain power distribution across multiple sub-bands, more consistent with FRB signals; conversely, a lower mean suggests a more cellular RFI signal. Time-domain broadband entropy characterizes the uniformity of energy distribution in the time dimension of a broadband signal. Frequency-domain broadband entropy characterizes the uniformity of energy distribution in the frequency dimension of a broadband signal. Short-time Fourier transform entropy characterizes the uniformity of energy distribution in the joint time-frequency dimension of a broadband signal. Welch power spectrum broadband entropy characterizes the uniformity of power distribution in the statistical average frequency domain of a broadband signal.

[0045] Single entropy features may lead to misjudgments due to noise or special interference, while multiple entropy features complement each other from dimensions such as time domain, frequency domain, time-frequency joint, and smoothed frequency domain, which can comprehensively capture the essential differences between FRB and cellular RFI, significantly reducing the misjudgment rate. For example, regarding time domain broadband entropy, FRB energy is naturally dispersed with propagation, resulting in a high entropy value, while cellular RFI is mostly transient and has a low entropy value; regarding frequency domain broadband entropy, FRB covers a wide and uniform frequency range, resulting in a high entropy value, while cellular RFI is concentrated in the communication frequency band, resulting in a low entropy value; regarding short-time Fourier transform entropy, FRB has no concentrated region in the time-frequency plane, resulting in a high entropy value, while cellular RFI has obvious time-frequency clustering, resulting in a low entropy value.

[0046] Step S700: Based on the dispersion curve matching degree, generate a first decision value; based on the mean of the frequency domain narrowband entropy, generate a second decision value; based on the time domain broadband entropy and the short-time Fourier transform entropy, generate a third decision value; based on the mean of the Welch power spectrum narrowband entropy and the Welch power spectrum broadband entropy, generate a fourth decision value; and based on the frequency domain broadband entropy, generate a fifth decision value.

[0047] Among them, the first, second, third, fourth, and fifth decision values ​​are all binary output results, where 1 indicates that they conform to FRB characteristics and 0 indicates that they conform more to cellular RFI characteristics.

[0048] Step S900: The first decision value, the second decision value, the third decision value, the fourth decision value, and the fifth decision value are weighted and summed, and fast radio bursts or cellular radio frequency interference are identified based on the weighted summation result.

[0049] In this approach, different weights are assigned based on the physical reliability of different features, allowing the core features to dominate the final decision, while supplementary features are combined to reduce misjudgments. Dispersion curve matching degree is the core feature, and various entropy features are supplementary features.

[0050] In some embodiments, the theoretical time delay vector is determined as follows: for any candidate dispersion value, the theoretical time delay of each sub-band in the signal matrix is ​​determined; for any sub-band, time delay compensation is performed based on the theoretical time delay to obtain a compensated sub-band signal; the compensated sub-band signals are merged to obtain a merged signal, and the signal-to-noise ratio (SNR) of the merged signal is calculated; the maximum SNR among the various SNRs is identified, and the candidate dispersion value corresponding to the maximum SNR is taken as the target dispersion value; based on the target dispersion value, the target theoretical time delay of each sub-band is determined; based on the target theoretical time delay of each sub-band, the theoretical time delay vector of the signal matrix is ​​determined.

[0051] By setting the search range and search step size for the dispersion value, multiple candidate dispersion values ​​can be obtained. For any given candidate dispersion value... The theoretical time delay of each sub-band relative to the 0th sub-band (the highest frequency sub-band) can be calculated. Based on the theoretical time delay of each sub-band, the sub-bands are delayed to obtain time-delay compensated sub-band signals. The time-delay compensated sub-band signals are accumulated and combined into a single combined signal, and the signal-to-noise ratio of the combined signal is calculated.

[0052] By iterating through each candidate dispersion value, the signal-to-noise ratio (SNR) of the corresponding combined signal is obtained. The maximum value among the SNR values ​​is compared, and the candidate dispersion value corresponding to the maximum SNR is taken as the target dispersion value. Based on the target dispersion value, the target theoretical time delay of each sub-band relative to the 0th sub-band can be calculated. The target theoretical time delays of each sub-band are combined to obtain the theoretical time delay vector of the signal matrix.

[0053] In some embodiments, the estimated time delay vector is determined as follows: for any sub-band, the peak position index is corrected using the three-point parabolic interpolation method to obtain the corrected index; the corresponding estimated time delay is calculated based on the corrected index; and the estimated time delay vector of the signal matrix is ​​generated based on the estimated time delay of each sub-band.

[0054] For each sub-band signal, the index of the peak position is first corrected using three-point parabolic interpolation to make the peak position more accurate. Based on the corrected accurate index, the estimated time delay corresponding to that sub-band is calculated. The estimated time delays of all sub-bands are then combined to form the estimated time delay vector of the signal matrix.

[0055] In some embodiments, for any sub-band, the theoretical time delay is converted into the number of sampling points; the time delay of the sub-band is compensated based on the number of sampling points to obtain the compensated sub-band signal, and the length of each compensated sub-band signal is the same.

[0056] Specifically, for each sub-band signal, a uniform length is achieved through time delay compensation. For any given sub-band, its theoretical time delay is first converted into the number of sampling points; for example, the number of sampling points can be obtained based on the theoretical time delay and the sampling rate. Based on the number of sampling points, the sub-band signal is adjusted for "time alignment" to obtain the compensated sub-band signal. After compensation, the length of all sub-band signals is consistent, thereby eliminating the time difference between sub-bands and achieving synchronization.

[0057] In some embodiments, the specific steps for estimating the dispersion curve matching degree include:

[0058] Step S301: Perform modulus and mean downsampling on each sub-band of the signal matrix. The downsampling rate is D, and the downsampled signal matrix is ​​DXX:

[0059]

[0060] in, for matrix, This refers to the length of a single signal channel. Indicates the first The route carries a signal. .

[0061] Step S303: Set the search range for dispersion values. and step length And calculate the number of DM searches K:

[0062]

[0063] in, This is the rounding operator.

[0064] Step S305: Estimate the dispersion value of the signal by traversal method. .

[0065] First, initialize Reference signal-to-noise ratio Set it to a very small value, such as -1000dB;

[0066] For the k-th dispersion value :

[0067] 1) Calculate the M sub-band signals , The theoretical time delay relative to the 0th sub-band signal, which is the highest frequency sub-band. :

[0068]

[0069] in, The meaning is the same as above, and the unit is... ; The unit is .

[0070] 2) Convert the theoretical time delay of the sub-band into the number of sampling points. And based on this, the first The signal is delayed in the path to obtain the delayed signal. :

[0071]

[0072]

[0073] in, Indicates that there is One zero, = D is the downsampling factor, and fs is the sampling frequency of the input signal.

[0074] 3) Delay the M-channel signal Accumulate and merge into one signal :

[0075]

[0076] 4) Calculate the signal-to-noise ratio of the merged signal, including the following steps:

[0077] A) Find the maximum value of Stack, maxx, and its index, index;

[0078] B) Calculate the mean of the signal in the Stack, excluding maxx and several points before and after it:

[0079]

[0080] Where len_g is the length of the guard window, the purpose of which is to prevent signal spread from affecting the estimation accuracy of background noise;

[0081] C) Calculate candidate dispersion values Signal-to-noise ratio :

[0082]

[0083] D) Update and reference signal-to-noise ratio ;

[0084]

[0085]

[0086] If the current signal-to-noise ratio is greater than the reference signal-to-noise ratio, then update. and reference signal-to-noise ratio If not, then no update will be made.

[0087] Step S307, the result after traversal This is the target dispersion value of the signal at this moment. Based on the target dispersion value, the theoretical time delay of each sub-band can be obtained. .

[0088] Step S309: Calculate the signal dispersion curve matching degree. The specific steps are as follows:

[0089] 1) For each sub-band signal Perform smoothing filtering:

[0090]

[0091] Where W is the length of the smooth window, which can be 3-5.

[0092] 2) For any sub-band, iterate through it to find the maximum value. and the index in which it is located .

[0093] 3) Calculate the positional deviation of the maximum value using three-point parabolic interpolation. :

[0094]

[0095] 4) Correct the maximum position index and calculate the estimated delay of the sub-band:

[0096]

[0097] 5) Construct vectors from the estimated time delay and the target theoretical time delay of each channel to obtain the estimated time delay vector. and theoretical time delay vector .

[0098]

[0099]

[0100] 6) Calculation and The correlation coefficient is the desired degree of matching of the dispersion curve. .

[0101] In some embodiments, if the dispersion curve matching degree is greater than or equal to a first threshold, the first decision value is set to 1; if the mean value of the frequency domain narrowband entropy is greater than or equal to a second threshold, the second decision value is set to 1; if the time domain broadband entropy is greater than a third threshold and the short-time Fourier transform entropy is greater than a fourth threshold, the third decision value is set to 1; if the mean value of the Welch power spectrum narrowband entropy is greater than or equal to a fifth threshold or the Welch power spectrum broadband entropy is greater than or equal to a sixth threshold, the fourth decision value is set to 1; if the frequency domain broadband entropy is greater than a seventh threshold, the fifth decision value is set to 1.

[0102] By applying threshold judgments to different feature indicators, multiple independent decision results are generated: 1 indicates compliance with FRB characteristics, and 0 indicates a closer resemblance to cellular RFI characteristics. Combining these decision results for a comprehensive judgment reduces the false positive rate.

[0103] In some embodiments, the first threshold is 0.8, the second threshold is 2, the third threshold is 6, the fourth threshold is 5.5, the fifth threshold is 3, the sixth threshold is 4, and the seventh threshold is 5.

[0104] In some embodiments, a weighted summation is performed on the first decision value, the second decision value, the third decision value, the fourth decision value, and the fifth decision value, including: setting the weight of the first decision value as a first weight, setting the weight of the second decision value as a second weight, and setting the weights of the third decision value, the fourth decision value, and the fifth decision value as a third weight, wherein the first weight is greater than the second weight, and the second weight is greater than the third weight.

[0105] Among them, the core feature is the dispersion curve matching degree, which has the highest weight for the first decision value, ensuring that it plays a leading role in the decision-making process; the important supplementary feature is the mean of the narrowband entropy in the frequency domain, which has the next highest weight for the second decision value; other supplementary features include the broadband entropy in the time domain, the short-time Fourier transform entropy, the mean of the narrowband entropy of the Welch power spectrum, the broadband entropy of the Welch power spectrum, and the broadband entropy in the frequency domain, which are used to assist in the confirmation. Individually, they may be misjudged, and the corresponding third decision values ​​have a lower weight.

[0106] In some embodiments, the first weight is 0.5, the second weight is 0.2, and the third weight is 0.1.

[0107] In some embodiments, if the weighted summation result is greater than or equal to a first preset value, it is determined to be a fast radio burst; if the weighted summation result is less than or equal to a second preset value, it is determined to be cellular radio frequency interference.

[0108] The higher the score of the weighted sum, the more likely it is to be an FRB; the lower the score, the more likely it is to be a cellular RFI.

[0109] In some embodiments, the first setting value is 0.7, and the second setting value is 0.3.

[0110] In some embodiments, the temporal broadband entropy is extracted. The calculation steps are as follows:

[0111] 1) Divide the signal into Segments, each segment is [length missing] , and satisfy ;

[0112] 2) Calculate the energy of each signal segment. :

[0113]

[0114] in, , Subband signal or broadband signal .

[0115] 3) Calculate the signal energy of each segment and :

[0116]

[0117] 4) Calculate the time-domain energy entropy :

[0118]

[0119]

[0120] When x is a broadband signal, what we get is the time-domain broadband entropy. .

[0121] In some embodiments, frequency domain narrowband entropy is extracted. and frequency domain broadband entropy The calculation method is the same, including the following steps:

[0122] 1) Transform the signal from the time domain to the frequency domain using the Distributed Fourier Transform (DFT):

[0123]

[0124] 2) Calculate the frequency domain energy spectrum :

[0125]

[0126] Where "*" is the conjugate operator;

[0127] 3) Spectral energy normalization:

[0128]

[0129] 4) Calculate the frequency domain energy entropy :

[0130]

[0131] When x is a sub-band signal, what is obtained is the frequency domain narrowband entropy. When x is a broadband signal, what we obtain is the frequency domain broadband entropy. .

[0132] In some embodiments, the narrowband entropy of the Welch power spectrum is extracted. and Welch power spectrum broadband entropy The calculation method is the same, including the following steps:

[0133] 1) Perform framing and windowing processing on the signal, that is, divide the signal into M segments, each segment having a length of L, with adjacent frames overlapping by R points:

[0134]

[0135] in: , This is the floor operator; R is usually set to the floor value. (50% overlap) or (75% overlap); Window functions, such as the Hamming window, are used to reduce the negative impact of signal cutoff.

[0136] 2) Calculate the periodogram of each signal segment:

[0137]

[0138] in, The input signal sampling frequency is k; the number of frequency points is k. This is the energy compensation factor for the window function.

[0139] 3) The average Welch power spectrum was obtained. :

[0140]

[0141] 4) Frequency spectrum normalization:

[0142]

[0143] 5) Calculate the Welch power spectrum :

[0144]

[0145] When x is a subband signal, what we get is the Welch power spectrum narrowband entropy. When x is a broadband signal, what we obtain is the Welch power spectrum broadband entropy. .

[0146] In some embodiments, for broadband signals, the steps for calculating the short-time Fourier transform entropy are as follows:

[0147] 1) Broadband signal Framing involves dividing the signal into M frames and windowing, with each frame having a length of L and adjacent frames overlapping by R points.

[0148]

[0149] 2) Perform Discrete Fourier Transform on the windowed signal after frame segmentation:

[0150]

[0151] 3) Calculate the time-frequency energy matrix:

[0152]

[0153] 4) Calculate the normalized probability distribution:

[0154]

[0155] 5) Finally, calculate the short-time Fourier transform entropy of the broadband signal. :

[0156]

[0157] In some embodiments, combining the dispersion curve coordination degree and the above-mentioned multi-entropy characteristics, multi-layer logic decision is used to separate FRB and cellular RFI. The implementation steps are as follows:

[0158] 1) Core test based on dispersion curve matching degree, first decision value The calculation formula is as follows:

[0159]

[0160] Because FRBs have high dispersion characteristics, and the dispersion curves of cellular RFIs typically do not match theoretical cosmological models, based on the characteristic distribution of historical FRB and cellular RFI samples, 90% of FRB dispersion curves have a matching degree greater than 0.85, while 95% of cellular RFI dispersion curves have a matching degree less than 0.6. Referring to radio astronomy standards (such as the DM screening threshold for the CHIME telescope, which is typically 0.75-0.9), setting the first threshold to 0.9 or higher may result in missed detections. Therefore, the first threshold for checking the dispersion curve matching degree is set to 0.8. Furthermore, the first threshold needs to be fine-tuned based on the signal-to-noise ratio. A value of 1 indicates that the dispersion curve matching test has passed.

[0161] 2) The test is based on the mean of the narrowband entropy in the frequency domain, and the second decision value is used. The calculation formula is as follows:

[0162]

[0163] The frequency domain energy distribution of FRBs is more random (high entropy), while cellular interference (such as 5G signals) exhibits a regular spectrum. Therefore, testing the mean entropy of the narrowband frequency domain is used as an important supplementary feature, and the second threshold for testing the mean entropy of the narrowband frequency domain is set to 2. When dealing with low-frequency interference, the second threshold needs to be appropriately lowered. If... A value of 1 indicates that the test based on the mean of narrowband entropy in the frequency domain has passed.

[0164] 3) A joint test based on time-domain broadband entropy and short-time Fourier transform entropy is performed to identify periodic signals and signals with concentrated time-frequency energy, with a third decision value. The calculation formula is as follows:

[0165]

[0166] The time-domain energy of FRBs is concentrated and has a complex time-frequency distribution, while the time-frequency energy of periodic interference (such as LTE frames) is more regular. The third threshold for testing the time-domain broadband entropy is set to 6, and the fourth threshold for testing the short-time Fourier transform entropy is set to 5.5. If the third decision value... A value of 1 indicates that the joint test based on the time-domain broadband entropy and the short-time Fourier transform entropy has passed.

[0167] 4) Based on the joint test of the mean narrowband entropy of the Welch power spectrum and the broadband entropy of the Welch power spectrum, the fourth decision value is determined. The calculation formula is as follows:

[0168]

[0169] The power spectral entropy value of FRB is typically higher than that of cellular RFI. The fifth threshold for testing the mean narrowband entropy of the Welch power spectrum is set to 3, and the sixth threshold for testing the wideband entropy of the Welch power spectrum is set to 4. If the fourth decision value... A value of 1 indicates that the joint test based on the mean narrowband entropy of the Welch power spectrum and the broadband entropy of the Welch power spectrum has passed.

[0170] 5) To further verify the randomness of the signal spectrum, a test is performed based on the frequency domain broadband entropy, using the fourth decision value. The calculation formula is as follows:

[0171]

[0172] Set the seventh threshold for testing the broadband entropy in the frequency domain to 5. If the fourth decision value... A value of 1 indicates that the test based on frequency domain broadband entropy has passed.

[0173] 6) Using the above five tests, calculate the final score by weighted summation. :

[0174]

[0175] 7) Based on the final score To make the final judgment:

[0176]

[0177] If the weighted sum is greater than or equal to 0.7, for example, if the dispersion curve matching test and any two auxiliary feature tests pass, it is identified as a Fast Radio Burst (FRB). If the weighted sum is less than or equal to 0.3, for example, if the dispersion curve matching test fails and most auxiliary feature tests fail, it is identified as cellular radio frequency interference. If the weighted sum is greater than 0.3 and less than 0.7, further manual verification is required.

[0178] Additionally, please refer to Figure 2 , Figure 2 This is a schematic diagram of a fast radio burst and cellular radio frequency interference identification system provided in an embodiment of this application, as shown below. Figure 2As shown, the system includes: an acquisition module for acquiring multiple sub-band signals and constructing a signal matrix and a broadband signal based on the multiple sub-band signals, wherein the columns of the signal matrix correspond to sub-bands with different center frequencies, and the broadband signal is generated by superimposing the multiple sub-band signals; a generation module for generating a dispersion curve matching degree based on the estimated time delay vector and the theoretical time delay vector of the signal matrix; and an extraction module for extracting the frequency domain narrowband entropy and Welch power spectrum narrowband entropy for any sub-band in the signal matrix; determining the average frequency domain narrowband entropy based on the frequency domain narrowband entropy of each sub-band; determining the average Welch power spectrum narrowband entropy based on the Welch power spectrum narrowband entropy of each sub-band; and extracting the time domain broadband signal for the broadband signal. The system comprises: entropy, frequency domain broadband entropy, short-time Fourier transform entropy, and Welch power spectrum broadband entropy; a decision module, used to generate a first decision value based on the dispersion curve matching degree; a second decision value based on the mean of the frequency domain narrowband entropy; a third decision value based on the time domain broadband entropy and the short-time Fourier transform entropy; a fourth decision value based on the mean of the Welch power spectrum narrowband entropy and the Welch power spectrum broadband entropy; and a fifth decision value based on the frequency domain broadband entropy; and an identification module, used to perform a weighted summation of the first, second, third, fourth, and fifth decision values, and to identify fast radio bursts or cellular radio frequency interference based on the weighted summation result.

[0179] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0180] The fast radio burst and cellular radio frequency interference identification system in this embodiment is presented in the form of functional units. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0181] Please see Figure 3 , Figure 3 This application provides a schematic diagram of the structure of a computer device, as shown in the embodiment of the present application. Figure 3As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 3 Take a processor 10 as an example.

[0182] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0183] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0184] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0185] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0186] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0187] This application also provides a computer-readable storage medium. The methods described in this application can be implemented in hardware or firmware, or implemented as recordable on a storage medium, or implemented as computer code downloaded over a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and subsequently stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the methods shown in the above embodiments are implemented.

[0188] This application provides a computer program product including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the method of any embodiment of this application.

[0189] The systems or modules described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0190] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0191] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0192] This application is described with reference to flowchart illustrations and / or block diagrams of methods, systems, and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0193] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0194] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0195] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0196] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on its differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0197] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

[0198] Although embodiments of this application have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of this application, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for identifying a fast radio burst and a cellular radio frequency interference, characterized in that, The method comprises: acquiring a plurality of sub-band signals, and constructing a signal matrix and a wide-band signal based on the plurality of sub-band signals, wherein columns of the signal matrix correspond to sub-bands of different center frequencies, and the wide-band signal is generated by superimposing the plurality of sub-band signals; generating a dispersion curve matching degree based on an estimated time delay vector and a theoretical time delay vector of the signal matrix; extracting a frequency domain narrow-band entropy and a Welch power spectrum narrow-band entropy of each sub-band in the signal matrix respectively, determining a frequency domain narrow-band entropy average based on the frequency domain narrow-band entropy of each sub-band, determining a Welch power spectrum narrow-band entropy average based on the Welch power spectrum narrow-band entropy of each sub-band, and extracting a time domain wide-band entropy, a frequency domain wide-band entropy, a short-time Fourier transform entropy, and a Welch power spectrum wide-band entropy of the wide-band signal respectively; generating a first decision value based on the dispersion curve matching degree, a second decision value based on the frequency domain narrow-band entropy average, a third decision value based on the time domain wide-band entropy and the short-time Fourier transform entropy, a fourth decision value based on the Welch power spectrum narrow-band entropy average and the Welch power spectrum wide-band entropy, and a fifth decision value based on the frequency domain wide-band entropy; performing weighted summation on the first decision value, the second decision value, the third decision value, the fourth decision value, and the fifth decision value, and identifying a fast radio burst or a cellular radio frequency interference based on a result of the weighted summation.

2. The method of claim 1, wherein, The theoretical time delay vector is determined in the following manner: for any candidate dispersion value, determining a theoretical time delay of each sub-band in the signal matrix, and for any sub-band, performing time delay compensation based on the theoretical time delay to obtain a compensated sub-band signal; merging the compensated sub-band signals to obtain a merged signal, and calculating a signal-to-noise ratio of the merged signal; identifying a maximum signal-to-noise ratio in the signal-to-noise ratios, and taking a candidate dispersion value corresponding to the maximum signal-to-noise ratio as a target dispersion value; based on the target dispersion value, determining a target theoretical time delay of each sub-band; based on the target theoretical time delay of each sub-band, determining the theoretical time delay vector of the signal matrix.

3. The method according to claim 1 or 2, characterized in that, The estimated time delay vector is determined in the following manner: for any sub-band, correcting a peak position index using a three-point parabolic interpolation method to obtain a corrected index, and calculating an estimated time delay of the sub-band according to the corrected index; based on the estimated time delay of each sub-band, generating the estimated time delay vector of the signal matrix.

4. The method of claim 2, wherein, The method further comprises: for any sub-band, converting the theoretical time delay into a number of sampling points; based on the number of sampling points, performing time delay compensation on the sub-band to obtain a compensated sub-band signal, and the lengths of the compensated sub-band signals are the same.

5. The method of claim 1, wherein, The method further comprises: if the dispersion curve matching degree is greater than or equal to a first threshold value, setting the first decision value as 1; if the frequency domain narrowband entropy average is greater than or equal to a second threshold value, setting the second decision value as 1; if the time domain wideband entropy is greater than a third threshold value and the short-time Fourier transform entropy is greater than a fourth threshold value, setting the third decision value as 1; if the Welch power spectrum narrowband entropy average is greater than or equal to a fifth threshold value or the Welch power spectrum wideband entropy is greater than or equal to a sixth threshold value, setting the fourth decision value as 1; and if the frequency domain wideband entropy is greater than a seventh threshold value, setting the fifth decision value as 1.

6. The method of claim 5, wherein, The first threshold value is 0.8, the second threshold value is 2, the third threshold value is 6, the fourth threshold value is 5.5, the fifth threshold value is 3, the sixth threshold value is 4, and the seventh threshold value is 5.

7. The method according to claim 5 or 6, characterized in that, The weighted sum of the first decision value, the second decision value, the third decision value, the fourth decision value, and the fifth decision value comprises: the weight of the first decision value is set as a first weight, the weight of the second decision value is set as a second weight, and the weights of the third decision value, the fourth decision value, and the fifth decision value are all set as a third weight, wherein the first weight is greater than the second weight, and the second weight is greater than the third weight.

8. The method of claim 7, wherein, The first weight is 0.5, the second weight is 0.2, and the third weight is 0.

1.

9. The method of claim 8, wherein, If the result of the weighted sum is greater than or equal to a first set value, the event is determined as a fast radio burst; and if the result of the weighted sum is less than or equal to a second set value, the event is determined as a cellular radio frequency interference.

10. The method of claim 9, wherein, The first set value is 0.7, and the second set value is 0.3.

Images