Radiation source identification method and device based on signal structure characteristics, equipment and medium

Abstract

The application discloses a radiation source identification method and device based on signal structure characteristics, equipment and medium, belongs to the technical field of electromagnetic signal analysis, the method is aimed at different system burst signal to complete characteristic word matching, and selects the signal segment based on the matching result, carries out domain transformation processing, generates the corresponding domain transformation result; extract the best sampling point of the signal, and complete the structure characteristic extraction based on the sparse feature difference processing mode; according to the structure characteristic, the individual identification of radiation source is completed. The application can effectively distinguish the communication radiation source individual by extracting the signal structure characteristics, analyzing and comparing the common points of the structure characteristics between the same radiation source individuals and the differences of the structure characteristics between different radiation source individuals, which is a key supplement to the existing radiation source individual identification method.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of electromagnetic signal analysis, and particularly relates to a radiation source identification method and device based on signal structure characteristics, equipment and medium. BACKGROUND

[0002] The communication signal in the electromagnetic field will be endowed with different characteristics in the generation, transmission and propagation stages, such as code stream characteristics, frame format characteristics, modulation characteristics and waveform characteristics introduced in the signal generation process; individual subtle characteristics formed due to the difference in nonlinearity of different hardware in the transmission stage. Therefore, when researchers carry out communication signal system discrimination or individual identification analysis, most of them carry out research based on the extraction of different dimensional characteristics of the signal. For example, when identifying the modulation mode, based on the basic principle of communication signal modulation, amplitude shift keying, frequency shift keying and phase shift keying respectively use the amplitude, frequency and phase of the sine wave to transmit the digital baseband signal, so each kind of modulated signal shows different characteristics in the time domain, frequency domain, instantaneous frequency, instantaneous phase and constellation diagram, and the multi-dimensional modulation characteristics of the signal can be comprehensively utilized for modulation mode discrimination; when identifying a specific signal system, based on the design characteristics of the preamble, frequency band occupation and other aspects of different specific signals of different systems, the special texture characteristics shown by the signal in different dimensions such as time-frequency transform are utilized for discrimination; when identifying the individual of the radiation source, the subtle difference characteristics shown on the signal due to the difference in signal transmission environment or the characteristics of the device elements of the transmitter are mainly utilized, and these small differences are shown on the radiation source signal in the form of phase noise and dispersion, frequency stability, parasitic modulation, transient working characteristics and the like. The commonly used individual identification method of the communication radiation source mainly extracts the unique characteristics of each communication radiation source through the acquisition and high-precision measurement of various subtle characteristics of the radiation source signal, so as to achieve the purpose of accurately identifying and tracking the communication radiation source.

[0003] The existing individual identification method of the radiation source mainly extracts the subtle characteristics of the transient or steady-state part of the signal. SUMMARY

[0004] The present application aims to overcome the defects of the prior art and provide a radiation source identification method, device, equipment and medium based on signal structure characteristics, which realizes the accurate extraction of signal structure characteristics such as pre-carrier, pre-symbol, post-carrier and post-symbol based on the state distribution characteristics of different signal samples, combined with domain transformation and sparse feature extraction approach, and provides signal classification characteristic element support for subsequent radiation source individual identification analysis.

[0005] The purpose of the present application is achieved by the following technical solutions:

[0006] A signal structure characteristic extraction method, the method comprising:

[0007] In response to the acquired signal segment, domain transformation processing is performed on the signal segment to obtain a domain transformation result of the signal segment;

[0008] An optimal sampling point of the signal segment is obtained according to the domain transformation result, and parameter information of the signal segment is recorded, the parameter information including a statistical total length, a sliding window length and a single statistical window length, and a corresponding statistical signal sequence is obtained according to the statistical total length and the single statistical window length;

[0009] A sequence number of a maximum value of the signal segment is obtained as a first sequence number, and a sequence of optimal sampling points of an instantaneous frequency signal is obtained with the first sequence number as a starting point and a sliding window length as an interval;

[0010] Based on the sequence of optimal sampling points, a sparse feature difference nested processing method is used to obtain a structural feature of the signal, the structural feature including a pre / post symbol starting position sequence number, a pre / post carrier end position sequence number and a pre / post carrier starting position sequence number;

[0011] According to the signal structural feature, individual identification of a radiation source is completed.

[0012] Further, the method further includes:

[0013] Before extracting the signal structural feature, prior analysis of the signal is performed, including burst signal detection, signal state distribution feature analysis and modulation mode identification, to obtain prior knowledge of the signal, and the prior knowledge of the signal specifically includes:

[0014] A burst signal detection algorithm is used to complete burst signal detection to obtain each complete burst signal;

[0015] State distribution feature analysis is performed on each burst signal respectively to determine whether the burst signal has a specific structural feature, and then the type of signal structural feature to be extracted is determined, the specific structural feature including a pre / post carrier, a pre / post symbol;

[0016] A baseband modulation mode of each burst signal is analyzed, and a domain transformation method is matched according to the baseband modulation mode.

[0017] Further, the domain transformation processing on the signal segment specifically includes:

[0018] A local carrier synchronization sequence of a preset length is intercepted, the local carrier synchronization sequence including a carrier and a symbol, and a local symbol synchronization sequence of a preset length is intercepted, the local symbol synchronization sequence including a symbol and a feature word;

[0019] According to the processing window length of the signal segment and the signal sampling rate, the instantaneous frequency of the signal segment, the instantaneous frequency of the local carrier synchronization sequence and the instantaneous frequency of the local symbol synchronization sequence are obtained;

[0020] The carrier synchronization and symbol synchronization processing are carried out, the instantaneous frequency of the signal segment is cross-correlated with the instantaneous frequency of the local carrier synchronization sequence and the instantaneous frequency of the local symbol synchronization sequence, the characteristic word matching is completed, the signal sequence number exceeding the peak threshold during the carrier synchronization processing and the signal sequence number exceeding the peak threshold during the symbol synchronization processing are recorded, and the signal sequence number exceeding the peak threshold during the symbol synchronization processing is taken as the preamble symbol end position sequence number;

[0021] According to the different modulation modes of the signal segments, the signal domain transformation method is selected.

[0022] Further, the sparse feature difference nested processing method for obtaining the structural feature of the signal specifically includes:

[0023] The optimal sampling point sequence is subjected to difference processing, the difference result amplitude absolute value is obtained, and the threshold value is compared, and then the structural feature of the signal is calculated according to the preselected nested layer number and the parameter information.

[0024] Further, the method selects the required nested layer number according to the signal distribution characteristics of the preposed carrier.

[0025] Further, the individual identification of the radiation source according to the signal structural feature includes:

[0026] The statistical information difference of the signal structural feature is utilized to perform the individual identification of the radiation source.

[0027] Further, the individual identification of the radiation source according to the signal structural feature includes:

[0028] Based on the original signal of the signal structural feature, deep learning is performed, so as to extract the individual difference of the radiation source and realize the individual identification of the radiation source.

[0029] In another aspect, the present application also provides a signal structural feature extraction device, which comprises:

[0030] A domain transformation module, which is responsive to the obtained signal segment, performs domain transformation processing on the signal segment to obtain a domain transformation result of the signal segment;

[0031] A parameter and statistical signal sequence calculation module calculates the optimal sampling points of the signal segment according to the domain transformation result, and records the parameter information of the signal segment, the parameter information including a total statistical length, a sliding window length and a single statistical window length, and calculates the corresponding statistical signal sequence according to the total statistical length and the single statistical window length;

[0032] An optimal sampling point sequence calculation module calculates the sequence number of the maximum value of the signal segment as a first sequence number, takes the first sequence number as the starting point of the optimal sampling points, and calculates the optimal sampling point sequence of the instantaneous frequency signal with an interval of the sliding window length.

[0033] A structural feature extraction module calculates the structural features of the signal based on the optimal sampling point sequence by using a sparse feature difference nested processing method, the structural features including the starting position sequence number of the pre / post code symbol, the ending position sequence number of the pre / post carrier and the starting position sequence number of the pre / post carrier.

[0034] In another aspect, the present application also provides a computer device, which comprises a processor and a memory, and the memory stores a computer program, the computer program is loaded and executed by the processor to realize any one of the above-mentioned radiation source identification methods based on signal structural features.

[0035] In another aspect, the present application also provides a computer readable storage medium, which stores a computer program, the computer program is loaded and executed by the processor to realize any one of the above-mentioned radiation source identification methods based on signal structural features.

[0036] The present application has the following advantages:

[0037] (1) The existing radiation source individual identification method is mainly based on the subtle features of the transient or steady-state part of the signal. The present application proposes to use the signal structural features that contain the semantic description rule information of the communication radio station as the basis for radiation source individual identification, which expands the mode of radiation source individual identification.

[0038] (2) The existing signal feature extraction method is mainly realized by two ways: one is to calculate the statistical parameter information based on the original signal to complete the feature expression, and the other is to input the original signal into a deep learning model to extract its deep features. The above-mentioned feature extraction methods are easily disturbed by noise, and the method of simply relying on deep learning to extract signal features has poor interpretability and poor robustness. The signal structural feature extraction method proposed in the present application is realized based on heterogeneous multi-dimensional information fusion, has high feature extraction accuracy, can adapt to signal structural feature extraction of as low as one chip length, and taking the signal structural features as the input of the radiation source individual intelligent identification model can effectively improve the robustness of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 Schematic diagram of signal structure feature extraction based on heterogeneous multi-dimensional information fusion according to an embodiment of the present invention;

[0040] Figure 2 This is a diagram of a typical shortwave / ultra-shortwave radio communication signal structure including a pre-carrier / symbol and a post-carrier / symbol;

[0041] Figure 3 It is a flowchart of the signal prior analysis process;

[0042] Figure 4 This is the principle block diagram of sparse signal differential nested processing;

[0043] Figure 5 This is a structural block diagram of a radiation source identification device based on signal structure characteristics according to an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The following describes the embodiments of the present invention through specific examples. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. The details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other unless they conflict.

[0045] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of the present invention.

[0046] Most existing methods for identifying individual radiation sources are based on extracting subtle features of the transient or steady-state parts of the signal.

[0047] In order to solve the above technical problems, the following embodiments of the present invention are proposed, including a method, device, equipment, and medium for identifying a radiation source based on signal structure characteristics.

[0048] Example 1

[0049] This embodiment proposes a signal structure feature extraction method based on heterogeneous multi-dimensional information fusion. It is mainly based on the state distribution characteristics of different signal samples, combined with domain transformation and sparse feature differential nested extraction method, to achieve accurate extraction of signal structure features such as pre-carrier, pre-code element, post-carrier, and post-code element, providing signal classification feature element support for subsequent individual radiation source identification and analysis.

[0050] Reference Figure 1 ,likeFigure 1 The signal structure feature extraction schematic diagram based on heterogeneous multi-dimensional information fusion of the embodiment is shown. The method comprises:

[0051] Step 1: Perform signal priori analysis, including burst signal detection, signal state distribution feature analysis and modulation mode identification, to obtain signal priori knowledge;

[0052] Step 2: Complete feature word matching for burst signals of different systems, and select signal segments based on the matching results, perform domain transformation processing, and generate corresponding domain transformation results;

[0053] Step 3: Extract the best sampling point of the signal, and complete structure feature extraction based on the sparse feature difference processing mode.

[0054] When the short wave / ultra-short wave communication radio station works, it will communicate with other radio stations regularly / irregularly according to its preset working mode, and the communication information carrier is electromagnetic signal. Due to the set working mode, the electromagnetic signal radiated by some communication radio stations has structure features such as pre-carrier / symbol and post-carrier / symbol, as shown in Figure 2 As shown in Figure 2 The typical short wave / ultra-short wave radio communication signal structure schematic diagram is shown. In most cases, the pre / post carrier is generally in the form of a single carrier, and the pre / post symbol is usually in the form of a multi-carrier with symmetry. By extracting the structure features contained in the intercepted electromagnetic signal, the fusion analysis of the signal structure features is helpful to realize the discrimination of the individual to which the communication radio station belongs. The following describes the implementation steps taking the baseband modulation of the FSK type signal as an example, which can be applied to other short wave / ultra-short wave communication signals with structure features such as single carrier pre / post carrier and symmetric multi-carrier pre / post symbol.

[0055] Referring to Figure 3 As shown in Figure 3 The signal priori analysis flowchart is shown. In step 1 provided by the embodiment, the specific process of signal priori analysis is as follows:

[0056] For the intercepted signal to be analyzed, first, the signal detection algorithm module is used to complete burst signal detection, and each complete burst signal is obtained. The starting position number of the burst signal obtained by signal detection is denoted as P start , and the ending position number is denoted as P endAfter that, the state distribution characteristics of each burst signal are analyzed, and it is roughly judged whether there are pre / post carrier, pre / post symbol and other structural characteristics, and the type of signal structure features to be extracted is determined, which can effectively improve the efficiency of subsequent structural feature extraction. At the same time, the modulation mode of each burst signal is determined, and the baseband modulation mode is classified, such as PSK, FSK or ASK, etc., so as to automatically match the most suitable domain transformation method in the subsequent process.

[0057] In step 2 provided by the embodiment, signal feature word matching and domain transformation processing are performed, and the specific implementation process is as follows:

[0058] If each burst signal has Figure 2 the complete signal structure as shown in the figure, P start can be taken as the rough estimation of the starting position number of the pre-carrier, and P end can be taken as the rough estimation of the ending position number of the post-carrier, and the signal segment division is performed based on the burst signal detection result, which can effectively reduce the redundancy and save the calculation time for obtaining the signal structure unit.

[0059] For short wave / ultra-short wave radio communication signals with pre-symbol and continuous 01 code stream feature words, first, a typical signal is selected, and a certain length of

carrier, symbol

symbol, feature word

[0060] Taking the baseband modulation as an example, the input signal to be tested is denoted as SigData(i)| i＝1,2,...,L , the signal processing window length is denoted as winL, the signal sampling rate is denoted as fs, PI≈3.1415926..., and the instantaneous frequency of the input signal to be tested is calculated by using formula (1), (2) and (3), denoted as fre(i)| i＝1,1,...,L-winL ,

[0061]

[0062] Phase(i)| i＝0,1,...,L-1 =arctan(SigData(i).*conj(SigData(i+1)))……(2)

[0063]

[0064] After that, the instantaneous frequency of the local carrier synchronization sequence and the local symbol synchronization sequence are calculated in the same way, and are recorded as CarSynFre and CodeSynFre respectively. Then, the carrier synchronization and symbol synchronization processing are carried out, that is, the cross-correlation calculation of fre with CarSynFre and CodeSynFre is carried out, the characteristic word matching is completed, the signal sequence number exceeding the peak threshold during the carrier synchronization processing is recorded as MaxIndex1, and the signal sequence number exceeding the peak threshold during the symbol synchronization processing is recorded as MaxIndex2. MaxIndex1 can be used as the coarse estimation of the pre-carrier end position sequence number. Since the characteristic word has uniqueness in the burst signal, MaxIndex2 can be considered as the pre-symbol end position sequence number, and the signal segment to be analyzed is selected based on the cross-correlation result, which can effectively shorten the time for calculating the pre-carrier / symbol start and end positions.

[0065] Finally, according to different modulation modes, the signal domain transformation method is selected, and the instantaneous frequency curve, instantaneous amplitude curve, instantaneous phase curve and the like of the corresponding signal segment are obtained, which provides support for subsequent accurate extraction of signal structure characteristics. Here, taking the FSK modulation system signal as an example for description, so the signal domain transformation method selected here is instantaneous frequency transformation.

[0066] In step 3 provided by the embodiment, signal structure feature extraction is carried out, and the specific implementation process is as follows:

[0067] First, the best sampling point is calculated based on the signal domain transformation result, and the total length is recorded as Num, the sliding window length is recorded as winSL, the single statistical window length is recorded as StaNum, and the statistical signal sequence SS is calculated as shown in formula (4):

[0068]

[0069] Wherein, mean(data, DL) represents the mean value of the signal sequence data with length DL, and abs(x-y) represents the absolute value of the difference between x and y.

[0070] Then, for the statistical signal sequence SS with length Num, the sequence number of the maximum value is calculated and recorded as MaxIndex. Taking MaxIndex as the starting point of the best sampling point, the interval is winSL, and the best sampling point sequence SS1 of the instantaneous frequency signal is calculated, and the sequence length is recorded as LL.

[0071] After that, the sparse feature difference nested processing method is used to calculate the pre-symbol start position, which is described with reference to Figure 4 For example, Figure 4The sparse signal difference nested processing principle diagram is shown, based on the best sampling point sequence SS1, difference processing, difference result amplitude absolute value calculation, threshold value comparison and other flow processing are carried out, the sequence number meeting the threshold requirement is recorded as nn; The same processing is carried out on nn, the sequence number meeting the threshold is recorded as ind, and the same processing is carried out on ind, the sequence number meeting the threshold is recorded as X, wherein the length of X is recorded as Num3; Wherein thd1, thd2, thd3 are configured according to the frequency value distribution, sparse characteristics and other of the preamble symbol. With the burst signal starting position P start As the reference, the preamble symbol starting position sequence number can be obtained by the following formula:

[0072] Code_ind1=nn[ind[X[Num3-1]+1]]×winSL

[0073] Since the preamble carrier wave is closely connected with the preamble symbol, Code_ind1 can be considered as the accurate value of the end position sequence number of the preamble carrier wave; Next, the accurate value of the starting position sequence number of the preamble carrier wave is calculated, and the sequence is solved based on SS1 as the basis, difference processing, difference result amplitude absolute value calculation are carried out, the sequence SS2 is obtained, the threshold value thd4 is set according to the size of the preamble carrier wave frequency value, the number and position sequence of the sampling points less than thd4 in the sequence SS2 are calculated, the number of sampling points meeting the condition is recorded as Num4, and the position sequence of the sampling points meeting the condition is recorded as nn1. Difference processing is carried out on nn1, and the data sequence ff1 is obtained. According to the single carrier characteristic of the preamble carrier wave, the number and position sequence of the sampling points with continuous data value of 1 in ff1 are counted, the number of sampling points meeting the condition is recorded as L, and the last position sequence meeting the condition is recorded as i. With the burst signal starting position P start As the reference, the accurate value of the starting position of the preamble carrier wave can be obtained by the following formula:

[0074] Car_ind1=(LL-nn1[i+1]-2)×winL

[0075] Generally, due to the similar signal distribution characteristics of the preamble carrier wave and the post carrier wave, and the similar signal distribution characteristics of the preamble symbol and the post symbol, the start and end position sequence numbers of the post carrier wave / symbol can be solved by referring to the start and end position sequence numbers of the preamble carrier wave / symbol. Only the way of taking and solving is from the end of the burst signal to the front.

[0076] The application is described in detail above in combination with the drawings, but it should be pointed out that the above examples are only preferred examples of the application and are not used to limit the application, and the application can have various changes and variations for those skilled in the art, such as the baseband modulation mode can be ASK, PSK, etc., when the signal structure feature is extracted, the instantaneous amplitude curve and the instantaneous phase curve are firstly calculated, and the analysis can be performed according to the characteristics of the signal instantaneous amplitude / instantaneous phase curve; the signal type can be other communication or radar radiation source signal. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the application shall be included in the scope of the claims of the application.

[0077] Embodiment 2

[0078] Referring to Figure 5 As Figure 5 shown is a structure block diagram of a radiation source identification device based on signal structure features in the embodiment, and the device specifically includes the following structures:

[0079] A domain transformation module, which performs domain transformation processing on the acquired signal segment to obtain a domain transformation result of the signal segment;

[0080] A parameter and statistical signal sequence calculation module, which calculates the optimal sampling point of the signal segment according to the domain transformation result, records the parameter information of the signal segment, the parameter information including the total statistical length, the sliding window length and the single statistical window length, and calculates the corresponding statistical signal sequence according to the total statistical length and the single statistical window length;

[0081] An optimal sampling point sequence calculation module, which calculates the sequence number of the maximum value of the signal segment as a first sequence number, takes the first sequence number as the starting point of the optimal sampling point, and calculates the optimal sampling point sequence of the instantaneous frequency signal with the interval being the sliding window length;

[0082] A structure feature extraction module, which calculates the structure features of the signal based on the optimal sampling point sequence by using a sparse feature difference nested processing method, and the structure features include the pre / post code symbol starting position sequence number, the pre / post carrier end position sequence number and the pre / post carrier starting position sequence number.

[0083] Embodiment 3

[0084] The preferred embodiment provides a computer device which can implement the steps in any of the embodiments of the signal structure feature based radiation source identification method provided in the embodiments of the present application, and thus can implement the beneficial effects of the signal structure feature based radiation source identification method provided in the embodiments of the present application, details of which are described in the foregoing embodiments and thus will not be described here.

[0085] Embodiment 4

[0086] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructions, or by controlling relevant hardware by the instructions, which can be stored in a computer readable storage medium and loaded and executed by a processor. For this purpose, the embodiments of the present application provide a storage medium having a plurality of instructions stored therein, which can be loaded by a processor to execute the steps in any of the embodiments of the signal structure feature based radiation source identification method provided in the embodiments of the present application.

[0087] The storage medium can include a read only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0088] Since the instructions stored in the storage medium can execute the steps in any of the embodiments of the signal structure feature based radiation source identification method provided in the embodiments of the present application, the beneficial effects that can be achieved by any of the signal structure feature based radiation source identification methods provided in the embodiments of the present application can be achieved, details of which are described in the foregoing embodiments and thus will not be described here.

[0089] The above only describes the preferred embodiments of the present application and is not intended to limit the present application, and any modifications, equivalent replacements and improvements made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A radiation source identification method based on signal structure characteristics, characterized in that: The method comprises: In response to the acquired radiation source signal segment, performing domain transformation processing on the signal segment to obtain a domain transformation result of the signal segment; Obtaining an optimal sampling point of the signal segment according to the domain transformation result, and recording parameter information of the signal segment, the parameter information including a total statistical length, a sliding window length, and a single statistical window length, and obtaining a corresponding statistical signal sequence according to the total statistical length and the single statistical window length; Obtaining a sequence number at which the maximum value of the statistical signal sequence is obtained as a first sequence number, using the first sequence number as an optimal sampling point starting point, and obtaining an optimal sampling point sequence for the instantaneous frequency signal at intervals equal to the sliding window length; Based on the optimal sampling point sequence, a sparse feature differential nesting processing method is used to obtain structural features of the signal, wherein the structural features include a preamble / postamble symbol starting position sequence number, a preamble / postamble carrier ending position sequence number, and a preamble / postamble carrier starting position sequence number. The sparse feature differential nesting processing method is used to obtain the structural features of the signal, specifically comprising: performing differential processing on the optimal sampling point sequence, obtaining an absolute value of the amplitude of the differential result, and comparing the amplitude with a threshold value, and then calculating the structural features of the signal based on a preselected number of nesting layers and the parameter information. According to the signal structure characteristics, individual identification of the radiation source is completed.

2. The radiation source identification method based on signal structure characteristics according to claim 1, characterized in that: The method further comprises: Before extracting signal structure features, perform signal prior analysis, including burst signal detection, signal state distribution characteristics analysis, and modulation mode identification, to obtain signal prior knowledge. Specifically, this includes: Use signal detection algorithm to complete burst signal detection and obtain each complete burst signal; Analyze the state distribution characteristics of each burst signal to determine whether the burst signal has specific structural features, and then determine the type of signal structural features that need to be extracted. Specific structural features include pre-carrier, post-carrier, pre-symbol, and post-symbol. The baseband modulation type of each burst signal is analyzed, and a domain transformation method is matched according to the baseband modulation type.

3. The radiation source identification method based on signal structure characteristics according to claim 1, characterized in that: The performing domain transformation processing on the signal segment specifically includes: intercepting a local carrier synchronization sequence of a preset length, the local carrier synchronization sequence including a carrier and a symbol, intercepting a local symbol synchronization sequence of a preset length, the local symbol synchronization sequence including a symbol and a signature; Calculating the instantaneous frequency of the signal segment, the instantaneous frequency of the local carrier synchronization sequence, and the instantaneous frequency of the local symbol synchronization sequence according to the processing window length of the signal segment and the signal sampling rate; Carry out carrier synchronization and symbol synchronization processing, perform cross-correlation calculation on the instantaneous frequency of the signal segment with the instantaneous frequency of the local carrier synchronization sequence and the instantaneous frequency of the local symbol synchronization sequence, complete feature word matching, record the signal sequence number that exceeds the peak threshold during carrier synchronization processing and the signal sequence number that exceeds the peak threshold during symbol synchronization processing, and use the signal sequence number that exceeds the peak threshold during symbol synchronization processing as the end position sequence number of the preamble symbol; The signal domain transformation method is selected according to the different modulation modes of the signal segments.

4. The radiation source identification method based on signal structure characteristics according to claim 1, characterized in that: The method selects the required number of nesting layers according to the signal distribution characteristics of the pre-carrier.

5. The radiation source identification method based on signal structure characteristics according to claim 1, characterized in that: The completing the individual identification of the radiation source according to the signal structure characteristics includes: The statistical information differences of signal structure characteristics are used to identify individual radiation sources.

6. The radiation source identification method based on signal structure characteristics according to claim 1, characterized in that: The completing the individual identification of the radiation source according to the signal structure characteristics includes: Deep learning is performed on the original signal based on the signal structure characteristics to extract the individual differences of the radiation source and realize the individual identification of the radiation source.

7. A radiation source identification device based on signal structure characteristics, characterized in that: The device comprises: a domain transformation module, wherein the domain transformation module performs domain transformation processing on the signal segment in response to the acquired signal segment to obtain a domain transformation result of the signal segment; a parameter and statistical signal sequence calculation module, which obtains the optimal sampling point of the signal segment based on the domain transformation result, records parameter information of the signal segment, the parameter information including the total statistical length, the sliding window length, and the single statistical window length, and obtains the corresponding statistical signal sequence based on the total statistical length and the single statistical window length; an optimal sampling point sequence calculation module, wherein the optimal sampling point sequence calculation module obtains a sequence number at which the maximum value of the statistical signal sequence is located as a first sequence number, uses the first sequence number as the optimal sampling point starting point, and obtains an optimal sampling point sequence for the instantaneous frequency signal at intervals equal to the sliding window length; A structural feature extraction module, wherein the structural feature extraction module uses a sparse feature differential nesting processing method to obtain the structural features of the signal based on the optimal sampling point sequence, and the structural features include the starting position sequence number of the pre / post code element, the ending position sequence number of the pre / post carrier, and the starting position sequence number of the pre / post carrier; the use of the sparse feature differential nesting processing method to obtain the structural features of the signal specifically includes: performing differential processing on the optimal sampling point sequence, obtaining the absolute value of the amplitude of the differential result and comparing it with the threshold value, and then calculating the structural features of the signal based on the preselected nesting layer number and the parameter information.

8. A computer device, characterized in that: The computer device includes a processor and a memory, wherein a computer program is stored in the memory, and the computer program is loaded and executed by the processor to implement the radiation source identification method based on signal structure characteristics as claimed in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that The storage medium stores a computer program, which is loaded and executed by a processor to implement the radiation source identification method based on signal structure characteristics according to any one of claims 1 to 6.

Images