Multifunction radar operating mode extraction method and apparatus

Abstract

This application relates to the field of multi-functional radar technology, and provides a method and apparatus for extracting the operating modes of multi-functional radar. The method, based on a large amount of pulse sequence sample data from a specified type of multi-functional radar, first performs cluster analysis on the carrier frequency, repetition interval, and pulse width parameters respectively. Then, based on the clustering results, it performs symbolization and parameterization processing on the pulse sequence sample data. Next, it performs frequent term mining on the symbolized pulse sequences and extracts pulse combination patterns using reduction rules, forming a radar word set through parameterization mapping. Finally, it performs serialization, frequent term mining, and reduction processing on the radar word set again to obtain radar word combination patterns, which are then parameterized to form a radar operating mode set. This achieves efficient extraction of multi-functional radar operating modes and can adapt to non-ideal data conditions such as pulse loss, false pulses, and measurement errors during the processing, thus improving extraction accuracy.

Description

Technical Field

[0001] This application relates to the field of multi-functional radar technology, and in particular to a method and apparatus for extracting the operating modes of a multi-functional radar. Background Technology

[0002] Multi-function radar (MFR) has multiple operating modes and can automatically schedule resources and switch operating modes according to the surrounding environment. It can transmit signals with different patterns and parameters to achieve multiple tasks such as multi-target search, acquisition, and tracking within the observation area. It is an important detection target of spaceborne passive detection systems.

[0003] Spaceborne passive detection systems can intercept the transmitted signals of multi-function radars and obtain the pulse sequences of individual multi-function radars through signal sorting and processing. Operating mode identification based on these pulse sequences is fundamental for threat assessment and jamming decisions for multi-function radars. However, achieving online identification requires prior information about the operating modes of multi-function radars. In the field of electronic warfare, multi-function radars are a crucial type of electronic equipment, and their operating modes and parameters are classified information. Therefore, it is necessary to analyze and mine the large amounts of data accumulated through passive detection to extract radar word templates and radar phrase templates for multi-function radars, and combine this with expert knowledge to form operating modes, thus supporting online analysis and identification of multi-function radar behavior.

[0004] Due to the flexible beam pointing and agile signal waveforms of multi-functional radars, non-cooperative spaceborne passive detection systems suffer from incomplete observations, and the received and processed multi-functional radar pulse sequences may contain missing pulses and measurement errors, making it challenging to accurately extract the complex operating modes of multi-functional radars. Summary of the Invention

[0005] In view of this, embodiments of this application provide a method and apparatus for extracting the operating mode of a multi-functional radar, in order to solve the technical problem in the prior art that it is difficult to accurately extract the operating mode of a multi-functional radar from passive detection data.

[0006] A first aspect of this application provides a method for extracting the operating modes of a multi-functional radar, comprising:

[0007] Acquire pulse sequence sample data of a specified type of multi-function radar from spaceborne passive detection data; the pulse sequence sample data includes N pulses, and the i-th pulse sequence sample data is... N is a positive integer greater than 2, and i is a positive integer greater than or equal to 1 and less than or equal to N; Let be the carrier frequency parameter of the i-th pulse sequence sample data. Let be the pulse width parameter of the i-th pulse sequence sample data. Let be the repetition interval parameter of the i-th pulse sequence sample data. and , Let i be the arrival time parameter of the i-th pulse sequence sample data;

[0008] The carrier frequency parameter, repetition interval parameter, and pulse width parameter are clustered separately. Based on the clustering results, the pulse sequence sample data is subjected to multi-parameter joint symbolization and parameterization processing to obtain the symbolized pulse sequence and the parameterized pulse sequence, and a mapping between the symbolized pulse sequence and the parameterized pulse sequence is established.

[0009] A multi-order symbol combination pattern chain is used to extract candidate symbol combination sequences from the symbolized pulse sequence, and a candidate radar word set is determined based on the candidate symbol combination sequences.

[0010] The candidate radar word set is symbolized again, and the symbolized candidate radar word set is parameterized and mapped to obtain the candidate working mode set.

[0011] The evidence theory approach is used to make decisions on the set of candidate operating modes, and the operating modes of the multi-functional radar are extracted.

[0012] A second aspect of this application provides a multi-functional radar operating mode extraction device, comprising:

[0013] The acquisition module is configured to acquire pulse sequence sample data of a specified type of multi-function radar from spaceborne passive detection data; the pulse sequence sample data includes N pulses, and the i-th pulse sequence sample data is... N is a positive integer greater than 2, and i is a positive integer greater than or equal to 1 and less than or equal to N; Let be the carrier frequency parameter of the i-th pulse sequence sample data. Let be the pulse width parameter of the i-th pulse sequence sample data. Let be the repetition interval parameter of the i-th pulse sequence sample data. and , Let i be the arrival time parameter of the i-th pulse sequence sample data;

[0014] The clustering module is configured to cluster the carrier frequency parameter, repetition interval parameter, and pulse width parameter respectively. Based on the clustering results, the pulse sequence sample data is subjected to multi-parameter joint symbolization and parameterization processing to obtain the symbolized pulse sequence and the parameterized pulse sequence, and a mapping is established between the symbolized pulse sequence and the parameterized pulse sequence.

[0015] The radar word extraction module is configured to use a multi-order symbol combination pattern chain to extract candidate symbol combination sequences from the symbolized pulse sequence, and determine a candidate radar word set based on the candidate symbol combination sequences.

[0016] The mapping module is configured to symbolize the candidate radar word set again and perform parameterized mapping on the symbolized candidate radar word set to obtain a candidate working mode set.

[0017] The operating mode extraction module is configured to use evidence theory methods to make decisions on the candidate operating mode set and extract the operating modes of the multi-functional radar.

[0018] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method.

[0019] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method.

[0020] The beneficial effects of this application embodiment compared with the prior art are as follows: This application embodiment, based on the accumulation of a large amount of pulse sequence sample data of a specified type of multi-functional radar, firstly performs cluster analysis on the carrier frequency, repetition interval, and pulse width parameters respectively, and then performs symbolization and parameterization processing on the pulse sequence sample data based on the clustering results. The three parameters of each pulse are symbolized as category numbers, and the cluster center value of each cluster represents the parameterization value of the pulse in this cluster. Next, frequent term mining is performed on the symbolized pulse sequence, and pulse combination patterns are extracted using reduction processing rules. A radar word set is formed through parameterization mapping. Finally, based on the radar word extraction results, the pulse sequence in the radar word set is subjected to secondary symbolization processing, and frequent term mining is performed again on the radar word symbolized pulse sequence. A radar word combination pattern is extracted using reduction processing rules, and a radar operating mode set is formed through parameterization mapping. This achieves efficient extraction of multi-functional radar operating modes, and can adapt to non-ideal data conditions such as pulse loss, false pulses, and measurement errors during the processing, thereby improving the extraction accuracy. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart illustrating a method for extracting the working mode of a multi-functional radar provided in an embodiment of this application.

[0023] Figure 2 This is a flowchart illustrating the method for extracting candidate symbol combination sequences from a symbolized pulse sequence using a multi-order symbol combination pattern chain, as provided in this application embodiment.

[0024] Figure 3 This is a flowchart illustrating a method for mining frequent itemsets in a multi-order symbolic combination chain set using a sliding window algorithm, as provided in an embodiment of this application.

[0025] Figure 4 This is a flowchart illustrating the method for reducing a set of candidate symbol combination chains provided in an embodiment of this application.

[0026] Figure 5 This is a flowchart illustrating the method for determining switching points using a sequence peak detection algorithm provided in an embodiment of this application.

[0027] Figure 6 This is a flowchart illustrating the method for obtaining a set of candidate radar words by parametric mapping of a reduced set of candidate symbol combination chains, as provided in an embodiment of this application.

[0028] Figure 7 This is a flowchart illustrating a method for periodic detection of a reduced candidate symbol combination chain set provided in an embodiment of this application.

[0029] Figure 8 This is a schematic diagram of a portion of the radar characters in a radar character set provided in an embodiment of this application.

[0030] Figure 9 This is a flowchart illustrating the method for determining a set of candidate working modes provided in an embodiment of this application.

[0031] Figure 10 This is a schematic diagram of radar word switching provided in an embodiment of this application.

[0032] Figure 11 This is a flowchart illustrating a method for making decisions on a set of candidate working modes using evidence theory, as provided in an embodiment of this application.

[0033] Figure 12 This is a schematic diagram of a multi-functional radar operating mode extraction device provided in an embodiment of this application.

[0034] Figure 13 This is a schematic diagram of the electronic device provided in the embodiments of this application. Detailed Implementation

[0035] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0036] The following will describe in detail, with reference to the accompanying drawings, a method and apparatus for extracting the working mode of a multi-functional radar according to an embodiment of this application.

[0037] Terminology Explanation:

[0038] Radar word: A fixed arrangement of a finite number of radar pulses that recurs stably in multi-functional radar pulses and is the basic unit for multi-functional radar pulse signal analysis.

[0039] Radar operating mode: A finite number of radar words are connected in series to represent one function or task of a multi-functional radar.

[0040] As mentioned above, due to the flexible beam pointing and agile signal waveform of multi-functional radar, non-cooperative spaceborne passive detection systems suffer from incomplete observations, and the multi-functional radar pulse sequence after reception and processing may have missing pulses and measurement errors, which makes it challenging to accurately extract the complex operating modes of multi-functional radar.

[0041] In the field of electronic warfare, the analysis of multi-functional radar signals mainly focuses on operating mode recognition and prediction. The main approach is based on the radar word templates of known multi-functional radars, converting individual multi-functional radar pulse sequences into symbolic radar word sequences through radar word extraction, and then identifying and predicting the operating mode. Most research in related technologies relies on publicly available radar word templates, radar phrase templates, and operating mode information from the "Mercury" radar for data generation and operating mode recognition. However, in practical applications, the detected multi-functional radars lack prior information such as radar word templates, which directly hinders the identification of multi-functional radar operating modes.

[0042] Based on the working mechanism and data analysis of multi-function radar, some typical shipborne multi-function radar signals exhibit obvious patterns. Their radar working mode information can be extracted from a large amount of data, thereby establishing a knowledge base or labeling the data to support subsequent radar working mode identification.

[0043] In view of this, this application provides a method for extracting the operating modes of a multi-functional radar. Based on a large amount of pulse sequence sample data from a specified type of multi-functional radar, the method first performs cluster analysis on the carrier frequency, repetition interval, and pulse width parameters. Then, based on the clustering results, the pulse sequence sample data is symbolized and parameterized. The three parameters of each pulse are symbolized as category numbers, and the cluster center value of each cluster represents the parameterized value of the pulse in that cluster. Next, frequent term mining is performed on the symbolized pulse sequences, and pulse combination patterns are extracted using reduction rules. A radar word set is formed through parameterized mapping. Finally, based on the radar word extraction results, the pulse sequences in the radar word set undergo secondary symbolization. Frequent term mining is performed again on the symbolized radar word pulse sequences, and radar word combination patterns are extracted using reduction rules. A radar operating mode set is formed through parameterized mapping, thereby achieving efficient extraction of the operating modes of a multi-functional radar. Furthermore, the method can adapt to non-ideal data conditions such as pulse loss, false pulses, and measurement errors during the processing, improving extraction accuracy.

[0044] Figure 1 This is a flowchart illustrating a method for extracting the operating mode of a multi-functional radar according to an embodiment of this application. Figure 1 As shown, the method includes the following steps:

[0045] In step S101, pulse sequence sample data of a specified type of multi-functional radar is obtained from the spaceborne passive detection data.

[0046] The pulse sequence sample data includes N pulses, and the i-th pulse sequence sample data is: N is a positive integer greater than 2, and i is a positive integer greater than or equal to 1 and less than or equal to N; Let be the carrier frequency parameter of the i-th pulse sequence sample data. Let be the pulse width parameter of the i-th pulse sequence sample data. Let be the repetition interval parameter of the i-th pulse sequence sample data. and , Let be the arrival time parameter of the i-th pulse sequence sample data.

[0047] In step S102, the carrier frequency parameter, repetition interval parameter, and pulse width parameter are clustered respectively. Based on the clustering results, the pulse sequence sample data is subjected to multi-parameter joint symbolization and parameterization processing to obtain the symbolized pulse sequence and the parameterized pulse sequence, and a mapping is established between the symbolized pulse sequence and the parameterized pulse sequence.

[0048] In step S103, a multi-order symbol combination pattern chain is used to extract candidate symbol combination sequences from the symbolized pulse sequence, and a candidate radar word set is determined based on the candidate symbol combination sequences.

[0049] In step S104, the candidate radar word set is symbolized again, and the symbolized candidate radar word set is parameterized and mapped to obtain the candidate working mode set.

[0050] In step S105, the evidence theory method is used to make a decision on the candidate operating mode set and extract the operating mode of the multi-functional radar.

[0051] In some embodiments of this application, the method may be executed by a server or by a terminal device with certain processing capabilities.

[0052] In some embodiments of this application, pulse sequence sample data of a specified type of multi-functional radar can be obtained from spaceborne passive detection data.

[0053] The pulse sequence sample data obtained from spaceborne passive detection data may include parameters such as Time of Arrival (TOA), Radio Frequency (RF), and Pulse Width (PW). The TOA of the radar pulse sequence can be first-order differentiated to obtain the Pulse Repetition Interval (PRI). The repetition interval parameter of the i-th pulse sequence sample data... At this point, the sample data of the i-th pulse sequence can be represented as For the first pulse sequence sample data, it can be used directly. Alternative That is, the sample data of the first pulse sequence can be represented as .

[0054] In some embodiments of this application, the carrier frequency parameter, repetition interval parameter, and pulse width parameter can be clustered separately. Based on the clustering results, the pulse sequence sample data can be subjected to multi-parameter joint symbolization and parameterization processing to obtain the symbolized pulse sequence and the parameterized pulse sequence, and a mapping between the symbolized pulse sequence and the parameterized pulse sequence can be established.

[0055] In some embodiments of this application, a multi-order symbol combination pattern chain can be used to extract candidate symbol combination sequences from the symbolized pulse sequence, and a candidate radar word set can be determined based on the candidate symbol combination sequences. Then, the candidate radar word set can be symbolized again, and the symbolized candidate radar word set can be parameterized and mapped to obtain a candidate operating mode set.

[0056] In some embodiments of this application, evidence theory methods are used to make decisions on a set of candidate operating modes to extract the operating mode of the multi-function radar. In one example, the Dempster combination rule (also known as DS evidence theory, or Dempster / Shafer evidence theory, which consists of the evidence theory proposed by Dempster and the imprecise reasoning theory further developed by Shafer) method can be used to make decisions on the set of candidate operating modes to determine the operating mode of the multi-function radar.

[0057] According to the technical solution provided in the embodiments of this application, based on the accumulation of a large amount of pulse sequence sample data of a specified type of multi-functional radar, firstly, cluster analysis is performed on the carrier frequency, repetition interval, and pulse width parameters respectively. Then, based on the clustering results, the pulse sequence sample data is symbolized and parameterized. The three parameters of each pulse are symbolized as category numbers, and the cluster center value of each cluster represents the parameterized value of the pulse in that cluster. Next, frequent term mining is performed on the symbolized pulse sequences, and pulse combination patterns are extracted using reduction processing rules. A radar word set is formed through parameterized mapping. Finally, based on the radar word extraction results, the pulse sequences in the radar word set are subjected to secondary symbolization processing. Frequent term mining is performed again on the symbolized radar word pulse sequences, and radar word combination patterns are extracted using reduction processing rules. A radar operating mode set is formed through parameterized mapping, thereby achieving efficient extraction of multi-functional radar operating modes. Moreover, it can adapt to non-ideal data conditions such as pulse loss, false pulses, and measurement errors during the processing, thus improving the extraction accuracy.

[0058] In some embodiments of this application, clustering the carrier frequency parameter, repetition interval parameter, and pulse width parameter can be performed separately to obtain... Individual carrier frequency parameter clusters, Clusters of repeat interval parameters and Each pulse width parameter is clustered; the cluster center, maximum value, and minimum value of each cluster are used as candidate parameter values ​​to obtain... Candidate carrier frequency parameter values, The candidate repetition interval parameter values ​​and One candidate pulse width parameter value.

[0059] In some implementations, the cluster center values ​​of each cluster can also be used as candidate parameter values ​​to obtain... Candidate carrier frequency parameter values, The candidate repetition interval parameter values ​​and One candidate pulse width parameter value.

[0060] in, , and All are positive integers.

[0061] In some embodiments of this application, multi-parameter joint symbolization and parameterization processing is performed on pulse sequence sample data based on clustering results to obtain symbolized pulse sequences and parameterized pulse sequences. Establishing a mapping between the symbolized pulse sequences and parameterized pulse sequences can be achieved by determining the first pulse sequence in the pulse sequence sample data. Carrier frequency symbolization value of each pulse Symbolic values ​​of repetition intervals And pulse width symbolic value ; Determine the first pulse sequence sample data Carrier frequency parameterization values ​​of each pulse Repeat interval parameterization value and pulse width parameterization value .

[0062] in, For the first The candidate carrier frequency symbolization value corresponding to each pulse. , Greater than or equal to 0 and less than or equal to ; For the first Symbolic values ​​of candidate repetition intervals corresponding to each pulse. , Greater than or equal to 0 and less than or equal to ; For the first The symbolic value of the candidate pulse width corresponding to each pulse. , Greater than or equal to 0 and less than or equal to .

[0063] For the first The parameterized value of the candidate carrier frequency corresponding to the nth pulse is the value of the nth pulse. Cluster center values ​​of cluster carrier frequency parameters; For the first The parameterized value of the candidate repetition interval corresponding to the nth pulse is the value of the nth pulse. Cluster center values ​​of the cluster repetition interval parameter; For the first The parameterized value of the candidate pulse width corresponding to the nth pulse is the value of the nth pulse. Cluster center value of cluster pulse width parameter.

[0064] The symbolized pulse sequence can be determined based on the symbolized values ​​of the carrier frequency, repetition interval, and pulse width of each pulse, and the parameterized pulse sequence can be determined based on the parameterized values ​​of the carrier frequency, repetition interval, and pulse width of each pulse.

[0065] Among them, the symbolized first Each pulse is The parameterized first Each pulse is ; For the symbolized first One pulse, For the parameterized first One pulse, Let be defined as the value on the left side of the equation, and map It is a one-to-one full-range response.

[0066] Figure 2 This is a flowchart illustrating the method for extracting candidate symbol combination sequences from a symbolized pulse sequence using a multi-order symbol combination pattern chain, as provided in an embodiment of this application. Figure 2 As shown, the method includes the following steps:

[0067] In step S201, the symbolized pulses in the symbolized pulse sequence are combined to obtain a multi-order symbol combination chain set.

[0068] Each symbol combination chain It consists of M symbolic pulse links. M is a symbol combination chain The length of M is a positive integer greater than 1 and less than N.

[0069] In step S202, the sliding window algorithm is used to mine frequent itemsets in the multi-order symbol combination chain set, and the mined frequent itemsets are used as candidate symbol combination chain sets.

[0070] In this context, each frequent item in the frequent itemset is a stable, recurring, fixed-length symbol combination sequence; a stable, recurring, fixed-length symbol combination sequence is any fixed-length symbol combination sequence in the symbol combination chain set whose recurrence count exceeds a preset threshold. In one example, multiple stable, recurring, fixed-length symbol combination sequences can be obtained based on different values ​​of M, M1 and M2, and different combinations from M1 to M2.

[0071] In step S203, the candidate symbol combination chain set is reduced to obtain the reduced candidate symbol combination chain set.

[0072] The reduction process can be used to remove redundant and duplicate subchains from the candidate symbol combination chain set.

[0073] In step S204, the reduced candidate symbol combination chain set is parameterized and mapped to obtain the candidate radar word set.

[0074] In some embodiments of this application, the symbolized pulses in the symbolized pulse sequence can be combined to obtain a set of multi-order symbol combination chains. Each symbol combination chain can be represented as follows: .

[0075] The sliding window algorithm can be used to mine stable, recurring symbol combination chains with fixed lengths from a set of multi-order symbol combination chains as frequent itemsets, and the mined frequent itemsets can be used as a candidate set of symbol combination chains. .

[0076] Can candidate symbol combination chain set The reduction process yields a reduced set of candidate symbol combination chains. Then, the reduced set of candidate symbol combination chains can be parametrically mapped to obtain a set of candidate radar characters.

[0077] Figure 3 This is a flowchart illustrating a method for mining frequent itemsets in a set of multi-order symbolic combination chains using a sliding window algorithm, as provided in an embodiment of this application. Figure 3 As shown, the method includes the following steps:

[0078] In step S301, selection is performed continuously starting from the first pulse. One pulse, to obtain the first Order symbol combination chain.

[0079] In step S302, slide one pulse sequentially backward to obtain all. Order symbol combination chain, determine the inclusion of all Order symbol combination chain A set of chained combinations of order symbols.

[0080] In step S303, the interval [2, N] is traversed. , confirm that it includes all A set of multi-order symbol combination chains.

[0081] In step S304, each of the multi-order symbol combination chain sets is counted. Frequency of occurrence of chain of order symbols.

[0082] In step S305, symbols with a frequency less than a preset frequency threshold are deleted from the multi-order symbol combination chain set. The order of symbol combination chains is used to obtain the set of candidate symbol combination chains.

[0083] In some embodiments of this application, when using the sliding window algorithm to mine frequent itemsets in a set of multi-order symbolic combination chains, continuous selection can be made starting from the first impulse. One pulse, to obtain the first A chain of sequence symbols. Then slide one pulse backward to obtain the second one. A chain of order symbols. Then slide sequentially until all are obtained. Order symbol combination chain, determine the inclusion of all Order symbol combination chain A set of chained combinations of order symbols.

[0084] Traverse the interval [2, N] , confirm that it includes all A set of multi-order symbolic combination chains. Statistical analysis of each element in the set of multi-order symbolic combination chains. Frequency of occurrence of chain of order symbols And delete symbols whose frequency is less than a preset frequency threshold from the multi-order symbol combination chain set. of By analyzing the chain of symbols of order, we can obtain the set of candidate chain of symbols. .

[0085] In practical applications, due to potential pulse loss, incomplete pulse sequences, and other issues, the extracted candidate symbol combination chain set may suffer from defects such as sequence fragmentation, sequence splicing, and sequence periodicity. Therefore, appropriate processing is required to obtain a more accurate symbol combination chain. In one example, reduction processing can overcome these defects.

[0086] Figure 4 This is a flowchart illustrating a method for reducing a set of candidate symbol combination chains provided in an embodiment of this application. Figure 4 As shown, the method includes the following steps:

[0087] In step S401, for any two candidate symbol combination chains of different lengths, reduction processing is performed starting from the shorter chain. If it is determined that the shorter chain is a continuous subchain of the longer chain, then the set of candidate symbol combination chains is... Short chains are removed until no two chains are reducible, resulting in a set of candidate symbol combinations after subchain reduction.

[0088] In step S402, switching point detection is performed on each candidate symbol combination chain in the candidate symbol combination chain set after subchain reduction to determine the switching point set.

[0089] Each switching point in the set of switching points All of these are peak points that satisfy the sequence peak detection algorithm.

[0090] In step S403, the candidate symbol combination chain set after subchain reduction is divided using the switching point set to obtain the divided candidate symbol combination chain, and the divided candidate symbol combination chain is added to the candidate symbol combination chain set to obtain the updated candidate symbol combination chain set.

[0091] In step S404, the updated candidate symbol combination chain set is... The process iteratively executes the following steps: determining frequent itemsets, subchain reduction, determining the set of switching points, re-determining the candidate symbol combination chains after segmentation, and updating the updated set of candidate symbol combination chains. This continues until, during the subchain reduction process, it is determined that no candidate symbol combination chain in the set is a subchain of any other candidate symbol combination chain, thus obtaining the reduced set of candidate symbol combination chains.

[0092] In some embodiments of this application, when performing reduction processing on the candidate symbol combination chain set, any two candidate symbol combination chains of different lengths can be processed. The reduction process begins with the shortest chain. If the shortest chain is determined to be a continuous subchain of the longer chain, then the candidate symbol combination chain set is used. Short chains are removed until no two chains are reducible, resulting in a set of candidate symbol combinations after subchain reduction. and All are sets less than or equal to The total number of candidate symbol combination chains is a positive integer, and .

[0093] For each candidate symbol combination chain in the set of candidate symbol combination chains after subchain reduction, the sequence peak detection algorithm can be used to detect the switching point, thereby determining the set of switching points.

[0094] In some embodiments of this application, the set of candidate symbol combination chains after subchain reduction can be divided using the set of switching points to obtain the divided candidate symbol combination chains. The segmented candidate symbol combination chain is then added to the candidate symbol combination chain set. This yields the updated set of candidate symbol combination chains.

[0095] The updated candidate symbol combination chain set can be used. The process iteratively executes the following steps: determining frequent itemsets, subchain reduction, determining the set of switching points, re-determining the candidate symbol combination chains after segmentation, and updating the updated set of candidate symbol combination chains. This continues until, during the subchain reduction process, it is determined that no candidate symbol combination chain in the set is a subchain of any other candidate symbol combination chain, thus obtaining the reduced set of candidate symbol combination chains.

[0096] Figure 5This is a flowchart illustrating the method for determining switching points using a sequence peak detection algorithm provided in an embodiment of this application. Figure 5 As shown, the method includes the following steps:

[0097] In step S501, the target parameterization values ​​of two adjacent pulses in the target candidate symbol combination chain are obtained respectively.

[0098] The target candidate symbol combination chain is any candidate symbol combination chain in the set of candidate symbol combination chains after sub-chain reduction, and the target parameterization values ​​include carrier frequency parameterization values, repetition interval parameterization values, and pulse width parameterization values.

[0099] In step S502, the first-order difference of the target parameterized values ​​of two adjacent pulses is determined, and the mean of the first-order difference of the target parameterized values ​​of the parameterized pulse sequence corresponding to the target candidate symbol combination chain is determined.

[0100] In step S503, in response to determining that the difference between the first-order difference and the mean of the first-order difference is greater than a preset difference threshold, the next pulse is determined as the switching point.

[0101] In some embodiments of this application, target parameterization values ​​of two adjacent pulses in the target candidate symbol combination chain are obtained respectively. The target candidate symbol combination chain can be any candidate symbol combination chain in the set of candidate symbol combination chains after sub-chain reduction, and the target parameterization values ​​include carrier frequency parameterization values, repetition interval parameterization values, and pulse width parameterization values.

[0102] The first-order difference between the target parameterized values ​​of two adjacent pulses can be determined, as well as the mean of the first-order difference between the target parameterized values ​​of the parameterized pulse sequence corresponding to the target candidate symbol combination chain. If the difference between the first-order difference and the mean of the first-order difference is greater than a preset difference threshold, then the latter pulse of the two adjacent pulses can be determined as the switching point.

[0103] In some implementations, the carrier frequency parameterization value, repetition interval parameterization value, and pulse width parameterization value in the target parameterization value can be processed sequentially according to a preset priority. In one example, it can be first determined whether the difference between the first-order difference of the pulse width parameterization value of the two adjacent pulses and the average first-order difference of the pulse width parameterization value of the target candidate symbol combination chain is greater than a first preset difference threshold. If so, it is further determined whether the difference between the first-order difference of the repetition interval parameterization value of the two adjacent pulses and the average first-order difference of the repetition interval parameterization value of the target candidate symbol combination chain is greater than a second preset difference threshold. If so, it is further determined whether the difference between the first-order difference of the carrier frequency parameterization value of the two adjacent pulses and the average first-order difference of the carrier frequency parameterization value of the target candidate symbol combination chain is greater than a third preset difference threshold. If so, the latter pulse of the two adjacent pulses is determined as the switching point.

[0104] Conversely, if the above judgment process results in a negative result once, it can be considered that neither of the two adjacent pulses is a switching point. In this case, the process can continue to traverse the next pair of adjacent pulses and repeat the above judgment process.

[0105] In some embodiments of this application, the values ​​of the first preset difference threshold, the second preset difference threshold, and the third preset difference threshold can be determined based on the radar type of the multi-function radar and the data characteristics of the pulse sequence sample data, and are not limited here. The values ​​of the first preset difference threshold, the second preset difference threshold, and the third preset difference threshold can be the same or different.

[0106] Figure 6 This is a flowchart illustrating the method for obtaining a candidate radar word set by parametric mapping of the reduced candidate symbol combination chain set, as provided in an embodiment of this application. Figure 6 As shown, the method includes the following steps:

[0107] In step S601, the reduced candidate symbol combination chain set is subjected to periodicity detection and deperiodicity processing.

[0108] In step S602, the reduced candidate symbol combination chain set obtained after periodic detection and deperiodization is mapped to the candidate radar word set of the model to which the specified multi-function radar belongs.

[0109] Among them, the candidate radar word set is used to represent the stable pulse sequence, and the mapping is a one-to-one full mapping.

[0110] In some embodiments of this application, the reduced candidate symbol combination chain set can be subjected to periodic detection and deperiodic processing, and then the reduced candidate symbol combination chain set obtained after periodic detection and deperiodic processing can be mapped to the candidate radar word set of the model to which the specified multi-function radar belongs.

[0111] Figure 7 This is a flowchart illustrating a method for periodic detection of a reduced set of candidate symbol combination chains provided in an embodiment of this application. Figure 7 As shown, the method includes the following steps:

[0112] In step S701, the symbolic pulse sequence to be detected is obtained from the reduced candidate symbol combination chain set.

[0113] Among them, the symbolic pulse sequence to be detected is the symbolic pulse sequence corresponding to any candidate symbol combination chain.

[0114] In step S702, the symbolic pulse sequence to be detected is used as a test pulse, and the sequence is traversed sequentially to find similar pulses.

[0115] In step S703, in response to determining that the first similar pulse has been found, the probe pulse and the first similar pulse constitute a detection period, and the same period is searched and recorded sequentially based on the detection period.

[0116] In step S704, in response to determining that no similar pulse has been found, the next sequence of symbolic pulses to be detected is obtained, and the next sequence of symbolic pulses to be detected is used as a new trial pulse to perform periodic detection again, until all the sequence of symbolic pulses to be detected in the candidate symbol combination chain set is traversed.

[0117] In some embodiments of this application, periodic detection of the reduced candidate symbol combination chain set can be performed by: obtaining a sequence of symbolized pulses to be detected from the reduced candidate symbol combination chain set; using this sequence as a probe pulse; and sequentially searching for similar pulses. If the first similar pulse is found, the probe pulse and the first similar pulse constitute a double-pulse detection period. Double-pulse detection is then performed sequentially in the reduced candidate symbol combination chain set to find and record the same period as the detection period formed by the double pulse. Furthermore, a complete period can be defined as an independent radar word.

[0118] In some implementations, after completing the double-pulse detection cycle traversal and probing, three-pulse and multi-pulse detection cycles can be constructed for traversal and probing. For example, the trial pulse and the second similar pulse found in the subsequent sequential traversal can form a three-pulse detection cycle, or the trial pulse and the nth similar pulse found in the subsequent sequential traversal can form a multi-pulse detection cycle, where n is a positive integer greater than 3.

[0119] On the other hand, if no similar pulse is found, the next sequence of symbolic pulses to be detected can be obtained, and the next sequence of symbolic pulses to be detected can be used as a new trial pulse to perform periodic detection again, until all the sequence of symbolic pulses to be detected in the candidate symbol combination chain set is traversed.

[0120] In some embodiments of this application, the periodization removal process for the reduced candidate symbol combination chain set can be performed by removing the common periods of the detection period from each periodized symbolized pulse sequence in the reduced candidate symbol combination chain set. Here, the periodized symbolized pulse sequence refers to the symbolized pulse sequence with a detection period in the reduced candidate symbol combination chain set.

[0121] Using the above method, the candidate radar word set can be obtained as follows: Any radar word in the candidate radar word set is . Figure 8 This is a schematic diagram of a portion of the radar characters in a radar character set provided in an embodiment of this application. For example... Figure 8As shown, the reduced candidate symbol combination chain [1,0,0]to[2,0,0] can be mapped to the first radar word RW1, the reduced candidate symbol combination chain [3,1,0]to[4,1,4]to[5,1,0] can be mapped to the second radar word RW2, and so on.

[0122] Figure 9 This is a flowchart illustrating the method for determining a set of candidate working modes provided in an embodiment of this application. Figure 9 As shown, the method includes the following steps:

[0123] In step S901, the candidate radar word set is symbolized again to obtain the symbolized candidate radar word set.

[0124] In step S902, frequent itemset mining and reduction processing are performed on the symbolized candidate radar word set to obtain the reduced candidate radar word set.

[0125] In step S903, the reduced candidate radar word set is parameterized and mapped to obtain the candidate working mode set.

[0126] In some embodiments of this application, the candidate radar word set can be symbolized again to obtain a symbolized candidate radar word set.

[0127] The parameter combination corresponding to the symbolized candidate radar word set can be expressed as: ; For the symbolized candidate radar characters, For assignment operator, For matrix functions, Mapping pulse symbolized sequences to radar word sets The inverse mapping, For parameterized representation symbols, To symbolize the symbol, For the first Candidate carrier frequency parameter values ​​for the pulse corresponding to each radar word For the first The candidate repetition interval parameter value of the pulse corresponding to each radar word. For the first Candidate pulse width parameter values ​​for the pulse corresponding to each radar word For the first Candidate carrier frequency parameter values ​​for each radar word. For the first The candidate repetition interval parameter value for each radar character. For the first Candidate pulse width parameter values ​​for each radar character. , This represents the number of radar characters in the symbolized candidate radar character set.

[0128] In the above-mentioned symbolized candidate radar word set, the three columns of the parameter combination matrix of the radar word represent the carrier frequency, repetition period, and pulse width of the pulse sequence corresponding to the radar word, respectively. Statistical analysis of each column can yield the inter-pulse modulation law and modulation parameters of the radar word, thereby further semantically describing the radar word.

[0129] In other words, it is possible to analyze radar word sets. The pulse sequences of various multi-functional radars are labeled and tagged with radar characters to obtain new radar characterized sequences. This allows for the analysis and discovery of the combination patterns of radar characters; among them, This is a new sequence of radar characters. Combinations of radar characters form the radar's operating mode, thereby enabling specific tasks such as search and tracking.

[0130] That is, for radar characterized pulse sequences, multi-order radar character sequences can be calculated and frequent term mining can be performed to extract the operating modes of multi-functional radars. ,in For the operating modes of the multi-functional radar, K is a positive integer greater than 1 and less than N. Then, a multi-order radar word chain is calculated based on the radar characterized radar word sequence. This multi-order radar word chain is then reduced again to remove redundancy and similar terms, thus obtaining the candidate operating mode set. .

[0131] Figure 10 This is a schematic diagram of radar word switching provided in an embodiment of this application. For example... Figure 10 As shown in the figure, this diagram illustrates a working mode consisting of two radar words, RW9 and RW15, which are periodically switched. Each radar word consists of four pulses. The horizontal axis represents the sequence symbol of the radar word, and the vertical axis is the index of the radar word, used to distinguish changes in the radar word sequence.

[0132] In some embodiments of this application, the candidate working mode set A work mode needs to be confirmed by experts before it can be approved for review. In one example, the DS evidence theory method can be used to analyze the set of candidate work modes. Make decisions.

[0133] Figure 11 This is a flowchart illustrating a method for making decisions on a set of candidate working modes using evidence theory, as provided in an embodiment of this application. Figure 11 As shown, the method includes the following steps:

[0134] In step S1101, the candidate working mode set is obtained. The basic probability allocation function is obtained from the expert review opinions.

[0135] in, It is a positive integer greater than 1.

[0136] In step S1102, the basic probability allocation function is fused using the Dempster combination rule to obtain the credibility of each working mode in the candidate working mode set.

[0137] In step S1103, the combination of working modes with a confidence level greater than a preset confidence level threshold is determined as the working mode of the multi-function radar.

[0138] In some embodiments of this application, when using evidence theory methods to make decisions on a set of candidate working modes, the set of candidate working modes can be obtained first. The basic probability allocation function is derived from the expert review opinions. ;in For the basic probability assignment function, For the set of candidate working modes, For any one of the candidate working modes in the set, It is a positive integer greater than 1.

[0139] Next, the Dempster combination rule can be used to fuse the basic probability assignment functions to obtain the confidence level of each working mode in the candidate working mode set. ,in For credibility.

[0140] Finally, it can be determined that the combination of operating modes with a confidence level greater than the preset confidence threshold is the operating mode of the multi-functional radar. ,in This is the operating mode for the multi-functional radar. The value of the threshold 'a' can be adjusted according to actual needs and is not restricted here. In one example, 'a' can be set to 0.7.

[0141] The technical solution provided in this application can decouple the parameter variation rules between the carrier frequency, repetition interval and pulse width of a multi-functional radar. It can symbolize the pulse sequence by single-parameter clustering and multi-parameter combination. At the same time, it can extract the radar word of the multi-functional radar by using a normalization processing method based on chain structure, and perform secondary symbolization and chain reduction on the pulse sequence based on the radar word to extract the candidate radar operating mode set. It can also use DS evidence theory to realize credible decision-making of operating modes based on expert knowledge, thus realizing accurate extraction of complex operating modes of multi-functional radar.

[0142] All of the above-mentioned optional technical solutions can be combined in any way to form the optional embodiments of this application, and will not be described in detail here.

[0143] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0144] Figure 12 This is a schematic diagram of a multi-functional radar operating mode extraction device provided in an embodiment of this application. Figure 12 As shown, the device includes:

[0145] The acquisition module 1201 is configured to acquire pulse sequence sample data of a specified type of multi-function radar from spaceborne passive detection data; the pulse sequence sample data includes N pulses, and the i-th pulse sequence sample data is... N is a positive integer greater than 2, and i is a positive integer greater than or equal to 1 and less than or equal to N; Let be the carrier frequency parameter of the i-th pulse sequence sample data. Let be the pulse width parameter of the i-th pulse sequence sample data. Let be the repetition interval parameter of the i-th pulse sequence sample data. and , Let be the arrival time parameter of the i-th pulse sequence sample data.

[0146] Clustering module 1202 is configured to cluster the carrier frequency parameter, repetition interval parameter and pulse width parameter respectively, and perform multi-parameter joint symbolization and parameterization processing on the pulse sequence sample data based on the clustering results to obtain the symbolized pulse sequence and the parameterized pulse sequence, and establish a mapping between the symbolized pulse sequence and the parameterized pulse sequence.

[0147] The radar word extraction module 1203 is configured to use a multi-order symbol combination pattern chain to extract candidate symbol combination sequences from the symbolized pulse sequence, and to determine a candidate radar word set based on the candidate symbol combination sequences.

[0148] The mapping module 1204 is configured to perform symbolization processing on the candidate radar word set again, and to perform parameterized mapping processing on the symbolized candidate radar word set to obtain a candidate working mode set.

[0149] The operating mode extraction module 1205 is configured to use evidence theory methods to make decisions on the candidate operating mode set and extract the operating modes of the multi-functional radar.

[0150] According to the technical solution provided in the embodiments of this application, based on the accumulation of a large amount of pulse sequence sample data of a specified type of multi-functional radar, firstly, cluster analysis is performed on the carrier frequency, repetition interval, and pulse width parameters respectively. Then, based on the clustering results, the pulse sequence sample data is symbolized and parameterized. The three parameters of each pulse are symbolized as category numbers, and the cluster center value of each cluster represents the parameterized value of the pulse in that cluster. Next, frequent term mining is performed on the symbolized pulse sequences, and pulse combination patterns are extracted using reduction processing rules. A radar word set is formed through parameterized mapping. Finally, based on the radar word extraction results, the pulse sequences in the radar word set are subjected to secondary symbolization processing. Frequent term mining is performed again on the symbolized radar word pulse sequences, and radar word combination patterns are extracted using reduction processing rules. A radar operating mode set is formed through parameterized mapping, thereby achieving efficient extraction of multi-functional radar operating modes. Moreover, it can adapt to non-ideal data conditions such as pulse loss, false pulses, and measurement errors during the processing, thus improving the extraction accuracy.

[0151] In some implementations, the carrier frequency parameter, repetition interval parameter, and pulse width parameter are clustered separately, including: clustering the carrier frequency parameter, repetition interval parameter, and pulse width parameter separately to obtain... Individual carrier frequency parameter clusters, Clusters of repeat interval parameters and Each pulse width parameter is clustered; the cluster center, maximum value, and minimum value of each cluster are used as candidate parameter values ​​to obtain... Candidate carrier frequency parameter values, The candidate repetition interval parameter values ​​and There are 10 candidate pulse width parameter values; among them... , and All are positive integers.

[0152] In some implementations, the pulse sequence sample data is subjected to multi-parameter joint symbolization and parameterization processing based on the clustering results to obtain symbolized pulse sequences and parameterized pulse sequences, and a mapping is established between the symbolized pulse sequences and parameterized pulse sequences, including: determining the first pulse sequence in the pulse sequence sample data. Carrier frequency symbolization value of each pulse Symbolic values ​​of repetition intervals And pulse width symbolic value ; Determine the first pulse sequence sample data Carrier frequency parameterization values ​​of each pulse Repeat interval parameterization value and pulse width parameterization value ;in, For the first The candidate carrier frequency symbolization value corresponding to each pulse. , Greater than or equal to 0 and less than or equal to ; For the first Symbolic values ​​of candidate repetition intervals corresponding to each pulse. , Greater than or equal to 0 and less than or equal to ; For the first The symbolic value of the candidate pulse width corresponding to each pulse. , Greater than or equal to 0 and less than or equal to ; For the first The parameterized value of the candidate carrier frequency corresponding to the nth pulse is the value of the nth pulse. Cluster center values ​​of cluster carrier frequency parameters; For the first The parameterized value of the candidate repetition interval corresponding to the nth pulse is the value of the nth pulse. Cluster center values ​​of the cluster repetition interval parameter; For the first The parameterized value of the candidate pulse width corresponding to the nth pulse is the value of the nth pulse. Cluster center values ​​for pulse width parameters; determining the symbolized pulse sequence based on the carrier frequency symbolization value, repetition interval symbolization value, and pulse width symbolization value of each pulse, and determining the parameterized pulse sequence based on the carrier frequency parameterization value, repetition interval parameterization value, and pulse width parameterization value of each pulse; wherein, the symbolized pulse sequence of the first pulse... Each pulse is The parameterized first Each pulse is ; For the symbolized first One pulse, For the parameterized first One pulse, Let be defined as the value on the left side of the equation, and map It is a one-to-one full-range response.

[0153] In some implementations, a multi-order symbol combination pattern chain is used to extract candidate symbol combination sequences from the symbolized pulse sequence, including: combining each symbolized pulse in the symbolized pulse sequence to obtain a set of multi-order symbol combination chains; each symbol combination chain... It consists of M symbolic pulse links. M is a symbol combination chain The length of M is a positive integer greater than 1 and less than N; the sliding window algorithm is used to mine frequent itemsets in the set of multi-order symbolic combination chains, and the mined frequent itemsets are used as candidate symbolic combination chain sets. Each frequent item in the frequent itemset is a stable, repetitive, and fixed-length symbol combination sequence; a stable, repetitive, and fixed-length symbol combination sequence is any fixed-length symbol combination sequence in the symbol combination chain set whose repetition count is greater than a preset threshold.

[0154] For the candidate symbol combination chain set The reduction process is performed to obtain a set of candidate symbol combination chains after reduction; the reduced set of candidate symbol combination chains is then parametrically mapped to obtain a set of candidate radar words.

[0155] In some implementations, a sliding window algorithm is used to mine frequent itemsets in a set of multi-order symbolic combination chains, and the mined frequent itemsets are used as a candidate set of symbolic combination chains. This includes: continuously selecting from the first impulse... One pulse, to obtain the first Order symbol combination chain; slide one pulse sequentially backward to obtain the complete chain. Order symbol combination chain, determine the inclusion of all Order symbol combination chain Set of chained combinations of order symbols; traversal , confirm that it includes all The set of multi-order symbolic combination chains; the statistical analysis of each multi-order symbolic combination chain set. Frequency of occurrence of chain of order symbols ; In the set of multi-order symbol combination chains, the deletion frequency is less than a preset frequency threshold. of Order symbol combination chains yield a set of candidate symbol combination chains. .

[0156] In some implementations, the candidate symbol combination chain set The reduction process includes: handling any combination of two candidate symbols of different lengths. The reduction process begins with the shortest chain. If the shortest chain is determined to be a continuous subchain of the longer chain, then the candidate symbol combination chain set is used. Short chains are removed until no two chains remain reducible, resulting in a set of candidate symbol combinations after subchain reduction; among them... and All are sets less than or equal to The total number of candidate symbol combination chains is a positive integer, and For each candidate symbol combination chain in the reduced candidate symbol combination chain set, a switching point detection is performed to determine the switching point set; each switching point in the switching point set... All are switching points that satisfy the sequence peak detection algorithm; the candidate symbol combination chain set after sub-chain reduction is obtained by using the switching point set to divide the sub-chain into segments. The segmented candidate symbol combination chain is then added to the candidate symbol combination chain set. This yields the updated set of candidate symbol combination chains; the updated set of candidate symbol combination chains... The process iteratively executes the following steps: determining frequent itemsets, subchain reduction, determining the set of switching points, re-determining the candidate symbol combination chains after segmentation, and updating the updated set of candidate symbol combination chains. This continues until, during the subchain reduction process, it is determined that no candidate symbol combination chain in the set is a subchain of any other candidate symbol combination chain, thus obtaining the reduced set of candidate symbol combination chains.

[0157] In some implementations, the sequence peak detection algorithm determines the switching point as follows: It acquires the target parameterized values ​​of two adjacent pulses in the target candidate symbol combination chain; the target candidate symbol combination chain is any candidate symbol combination chain in the set of candidate symbol combination chains after sub-chain reduction, and the target parameterized values ​​include carrier frequency parameterized values, repetition interval parameterized values, and pulse width parameterized values; it determines the first-order difference between the target parameterized values ​​of two adjacent pulses, and determines the mean of the first-order difference of the target parameterized values ​​of the parameterized pulse sequence corresponding to the target candidate symbol combination chain; in response to determining that the difference between the first-order difference and the mean of the first-order difference is greater than a preset difference threshold, it determines the next pulse as the switching point.

[0158] In some implementations, parameterized mapping is performed on the reduced candidate symbol combination chain set to obtain a candidate radar word set, including: performing periodic detection and deperiodization processing on the reduced candidate symbol combination chain set; mapping the reduced candidate symbol combination chain set obtained after periodic detection and deperiodization processing to a candidate radar word set of the model to which a specified multi-function radar belongs; the candidate radar word set is used to characterize a stably occurring pulse sequence, and the mapping is a one-to-one full mapping.

[0159] In some implementations, periodic detection is performed on the reduced candidate symbol combination chain set, including: obtaining a symbolic pulse sequence to be detected from the reduced candidate symbol combination chain set; the symbolic pulse sequence to be detected is the symbolic pulse sequence corresponding to any candidate symbol combination chain; using the symbolic pulse sequence to be detected as a probe pulse, sequentially traversing to find similar pulses; in response to determining that the first similar pulse has been found, using the probe pulse and the first similar pulse to form a detection period, and sequentially traversing to find and recording the same period based on the detection period; in response to determining that no similar pulse has been found, obtaining the next symbolic pulse sequence to be detected, and using the next symbolic pulse sequence to be detected as a new probe pulse to perform periodic detection again, until all the symbolic pulse sequences to be detected in the candidate symbol combination chain set have been traversed.

[0160] In some implementations, the reduced candidate symbol combination chain set is subjected to periodicity removal processing, including: removing the same period of the detection period from each periodized symbolized pulse sequence in the reduced candidate symbol combination chain set; wherein, the periodized symbolized pulse sequence is the symbolized pulse sequence with a detection period in the reduced candidate symbol combination chain set.

[0161] In some implementations, the candidate radar word set is Any radar word in the candidate radar word set is The candidate working mode set is determined as follows: the candidate radar word set is symbolized again to obtain the symbolized candidate radar word set; the parameter combination corresponding to the symbolized candidate radar word set is:

[0162] ; For the symbolized candidate radar characters, For assignment operator, For matrix functions, Mapping pulse symbolized sequences to radar word sets The inverse mapping, For parameterized representation symbols, To symbolize the symbol, For the first Candidate carrier frequency parameter values ​​for the pulse corresponding to each radar word For the first The candidate repetition interval parameter value of the pulse corresponding to each radar word. For the first Candidate pulse width parameter values ​​for the pulse corresponding to each radar word For the first Candidate carrier frequency parameter values ​​for each radar word. For the first The candidate repetition interval parameter value for each radar character. For the first Candidate pulse width parameter values ​​for each radar character. , The number of radar words in the symbolized candidate radar word set is given; frequent itemset mining and reduction processing are performed on the symbolized candidate radar word set to obtain the reduced candidate radar word set; parameterized mapping is performed on the reduced candidate radar word set to obtain the candidate working mode set.

[0163] In some implementations, evidence theory methods are used to make decisions on the set of candidate working modes, including:

[0164] Retrieving the set of candidate working modes The basic probability allocation function is derived from the expert review opinions. ;in For the basic probability assignment function, For the set of candidate working modes, It is any one of the candidate working modes in the set of working modes;

[0165] By fusing the basic probability assignment functions using the Dempster combination rule, the credibility of each working mode in the candidate working mode set is obtained. ,in To determine the credibility level, the combination of operating modes with a credibility level greater than a preset credibility threshold is identified as the operating mode of the multi-functional radar. ,in This is the operating mode for the multi-functional radar.

[0166] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0167] Figure 13 This is a schematic diagram of the electronic device provided in an embodiment of this application. For example... Figure 13 As shown, the electronic device 13 of this embodiment includes: a processor 1301, a memory 1302, and a computer program 1303 stored in the memory 1302 and executable on the processor 1301. When the processor 1301 executes the computer program 1303, it implements the steps in the various method embodiments described above. Alternatively, when the processor 1301 executes the computer program 1303, it implements the functions of each module / unit in the various device embodiments described above.

[0168] Electronic device 13 may be a desktop computer, laptop, handheld computer, cloud server, or other electronic device. Electronic device 13 may include, but is not limited to, processor 1301 and memory 1302. Those skilled in the art will understand that... Figure 13This is merely an example of electronic device 13 and does not constitute a limitation on electronic device 13. It may include more or fewer components than shown, or different components.

[0169] The processor 1301 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0170] The memory 1302 can be an internal storage unit of the electronic device 13, such as a hard disk or RAM of the electronic device 13. The memory 1302 can also be an external storage device of the electronic device 13, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, FlashCard, etc., equipped on the electronic device 13. The memory 1302 can also include both internal and external storage units of the electronic device 13. The memory 1302 is used to store computer programs and other programs and data required by the electronic device.

[0171] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0172] If an integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program may include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. A computer-readable medium may include: any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0173] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for extracting the operating modes of a multi-functional radar, characterized in that, include: Acquire pulse sequence sample data of a specified type of multi-functional radar from spaceborne passive detection data; The pulse sequence sample data includes N pulses, and the i-th pulse sequence sample data is... N is a positive integer greater than 2, and i is a positive integer greater than or equal to 1 and less than or equal to N; Let be the carrier frequency parameter of the i-th pulse sequence sample data. Let be the pulse width parameter of the i-th pulse sequence sample data. Let be the repetition interval parameter of the i-th pulse sequence sample data. and , Let i be the arrival time parameter of the i-th pulse sequence sample data; Clustering is performed on the carrier frequency parameter, the repetition interval parameter, and the pulse width parameter respectively. Based on the clustering results, the pulse sequence sample data is subjected to multi-parameter joint symbolization and parameterization processing to obtain the symbolized pulse sequence and the parameterized pulse sequence, and a mapping is established between the symbolized pulse sequence and the parameterized pulse sequence. The symbolized pulse sequence is extracted using a multi-order symbol combination pattern chain, and a candidate radar word set is determined based on the candidate symbol combination sequence. The candidate radar word set is symbolized again, and the symbolized candidate radar word set is parameterized and mapped to obtain the candidate working mode set. The evidence theory method is used to make a decision on the set of candidate operating modes, and the operating mode of the multi-functional radar is extracted.

2. The method according to claim 1, characterized in that, Clustering is performed on the carrier frequency parameter, repetition interval parameter, and pulse width parameter, including: Clustering is performed on the carrier frequency parameter, repetition interval parameter, and pulse width parameter respectively to obtain Individual carrier frequency parameter clusters, Clusters of repeat interval parameters and Clusters of pulse width parameters; Using the cluster center, maximum, and minimum values ​​of each cluster as candidate parameter values, we obtain... Candidate carrier frequency parameter values, The candidate repetition interval parameter values ​​and One candidate pulse width parameter value; in, , and All are positive integers.

3. The method according to claim 2, characterized in that, Based on the clustering results, the pulse sequence sample data undergoes multi-parameter joint symbolization and parameterization processing to obtain symbolized pulse sequences and parameterized pulse sequences. A mapping is then established between the symbolized and parameterized pulse sequences, including: Determine the first pulse sequence sample data Carrier frequency symbolization value of each pulse Symbolic values ​​of repetition intervals And pulse width symbolic value ; Determine the first pulse sequence sample data Carrier frequency parameterization values ​​of each pulse Repeat interval parameterization value and pulse width parameterization value ; in, For the first The candidate carrier frequency symbolization value corresponding to each pulse. , Greater than or equal to 0 and less than or equal to ; For the first Symbolic values ​​of candidate repetition intervals corresponding to each pulse. , Greater than or equal to 0 and less than or equal to ; For the first The symbolic value of the candidate pulse width corresponding to each pulse. , Greater than or equal to 0 and less than or equal to ; For the first The parameterized value of the candidate carrier frequency corresponding to the nth pulse is the value of the nth pulse. Cluster center values ​​of cluster carrier frequency parameters; For the first The parameterized value of the candidate repetition interval corresponding to the nth pulse is the value of the nth pulse. Cluster center values ​​of the cluster repetition interval parameter; For the first The parameterized value of the candidate pulse width corresponding to the nth pulse is the value of the nth pulse. Cluster center values ​​for cluster pulse width parameters; The symbolized pulse sequence is determined based on the carrier frequency symbolization value, repetition interval symbolization value, and pulse width symbolization value of each pulse, and the parameterized pulse sequence is determined based on the carrier frequency parameterization value, repetition interval parameterization value, and pulse width parameterization value of each pulse. Among them, the symbolized first Each pulse is The parameterized first Each pulse is ; For the symbolized first One pulse, For the parameterized first One pulse, Let be defined as the value on the left side of the equation, and map It is a one-to-one full-range response.

4. The method according to claim 1, characterized in that, The candidate symbol combination sequence is extracted from the symbolized pulse sequence using a multi-order symbol combination pattern chain, including: The symbolized pulses in the symbolized pulse sequence are combined to obtain a set of multi-order symbol combination chains; each symbol combination chain... It consists of M symbolic pulse links. M is the symbol combination chain. The length of M is a positive integer greater than 1 and less than N; The sliding window algorithm is used to mine frequent itemsets in the multi-order symbol combination chain set, and the mined frequent itemsets are used as candidate symbol combination chain sets. Each frequent item in the frequent item set is a stable, repetitive, and fixed-length symbol combination sequence; the stable, repetitive, and fixed-length symbol combination sequence is any fixed-length symbol combination sequence in the symbol combination chain set whose repetition count is greater than a preset threshold. For the candidate symbol combination chain set Perform reduction processing to obtain a reduced set of candidate symbol combination chains; The reduced candidate symbol combination chain set is parameterized and mapped to obtain the candidate radar word set.

5. The method according to claim 4, characterized in that, The sliding window algorithm is used to mine frequent itemsets in the set of multi-order symbol combination chains, and the mined frequent itemsets are used as candidate symbol combination chain sets, including: Select continuously starting from the first pulse. One pulse, to obtain the first Order symbol combination chain; Slide one pulse sequentially backward to obtain all. Order symbol combination chain, determine the inclusion of all Order symbol combination chain Set of chained combinations of order symbols; Traversal , confirm that it includes all A set of multi-order symbol combination chains; Statistically analyze each of the multi-order symbol combination chain sets Frequency of occurrence of chain of order symbols ; The deletion frequency in the multi-order symbol combination chain set is less than a preset frequency threshold. of Order symbol combination chain, to obtain the candidate symbol combination chain set. .

6. The method according to claim 4, characterized in that, For the candidate symbol combination chain set Regulations processing includes: For any two candidate symbol combinations of different lengths The reduction process begins with the shortest chain. If the shortest chain is determined to be a continuous subchain of the longer chain, then the process is performed using the candidate symbol combination chain set. Short chains are removed until no two chains remain reducible, resulting in a set of candidate symbol combinations after subchain reduction; among them... and All are sets less than or equal to The total number of candidate symbol combination chains is a positive integer, and ; For each candidate symbol combination chain in the reduced candidate symbol combination chain set, a switching point detection is performed to determine the switching point set; each switching point in the switching point set... All of these are switching points that satisfy the sequence peak detection algorithm; The candidate symbol combination chain set after the sub-chain reduction is divided using the set of switching points to obtain the divided candidate symbol combination chain. The segmented candidate symbol combination chain is then added to the candidate symbol combination chain set. This yields the updated set of candidate symbol combination chains; For the updated candidate symbol combination chain set The process iteratively executes the following operations: determining frequent itemsets, subchain reduction, determining switching point sets, re-determining the split candidate symbol combination chains, and updating the updated candidate symbol combination chain set, until it is determined during subchain reduction that no candidate symbol combination chain in the set is a subchain of any other candidate symbol combination chain, thus obtaining the reduced candidate symbol combination chain set.

7. The method according to claim 6, characterized in that, The sequence peak detection algorithm determines the switching point in the following manner: The target parameterization values ​​of two adjacent pulses in the target candidate symbol combination chain are obtained respectively; the target candidate symbol combination chain is any candidate symbol combination chain in the set of candidate symbol combination chains after the sub-chain reduction, and the target parameterization values ​​include carrier frequency parameterization value, repetition interval parameterization value and pulse width parameterization value; Determine the first-order difference of the target parameterized values ​​of the two adjacent pulses, and determine the mean of the first-order difference of the target parameterized values ​​of the parameterized pulse sequence corresponding to the target candidate symbol combination chain; In response to determining that the difference between the first-order difference and the mean of the first-order difference is greater than a preset difference threshold, the next pulse is determined as the switching point.

8. The method according to claim 4, characterized in that, The reduced candidate symbol combination chain set is parameterized and mapped to obtain a candidate radar word set, including: Perform periodicity detection and deperiodicity processing on the reduced candidate symbol combination chain set; The reduced candidate symbol combination chain set obtained after periodic detection and deperiodic processing is mapped to the candidate radar word set of the model to which the specified multi-function radar belongs; the candidate radar word set is used to characterize the stable pulse sequence, and the mapping is a one-to-one full mapping.

9. The method according to claim 8, characterized in that, Periodicity detection is performed on the reduced set of candidate symbol combination chains, including: From the reduced set of candidate symbol combination chains, the symbolized pulse sequence to be detected is obtained; the symbolized pulse sequence to be detected is the symbolized pulse sequence corresponding to any candidate symbol combination chain. Using the symbolic pulse sequence to be detected as a probe pulse, the system sequentially searches for similar pulses. In response to determining that the first similar pulse has been found, a detection period is formed by the probe pulse and the first similar pulse, and the same period is searched and recorded sequentially based on the detection period. In response to the determination that no similar pulse has been found, the next sequence of symbolic pulses to be detected is obtained, and the next sequence of symbolic pulses to be detected is used as a new probe pulse to perform periodic detection again, until all the sequence of symbolic pulses to be detected in the candidate symbol combination chain set is traversed.

10. The method according to claim 9, characterized in that, The reduced candidate symbol combination chain set is then deperiodized, including: For each periodized symbolized pulse sequence in the reduced candidate symbol combination chain set, remove the same period of its detection period; The periodicized symbolic pulse sequence is a symbolic pulse sequence with a detection period from the reduced candidate symbol combination chain set.

11. The method according to claim 8, characterized in that, The candidate radar word set is as follows Any radar word in the candidate radar word set is ; The set of candidate working modes is determined in the following manner: The candidate radar word set is symbolized again to obtain the symbolized candidate radar word set; Wherein, the parameter combination corresponding to the symbolized candidate radar word set is ; The symbolized set of candidate radar characters, For assignment operator, For matrix functions, Mapping pulse symbolized sequences to radar word sets The inverse mapping, For parameterized representation symbols, To symbolize the symbol, For the first Candidate carrier frequency parameter values ​​for the pulse corresponding to each radar word For the first The candidate repetition interval parameter value of the pulse corresponding to each radar word. For the first Candidate pulse width parameter values ​​for the pulse corresponding to each radar word For the first Candidate carrier frequency parameter values ​​for each radar word. For the first The candidate repetition interval parameter value for each radar character. For the first Candidate pulse width parameter values ​​for each radar character. , The number of radar characters in the symbolized candidate radar character set; Frequent itemset mining and reduction processing are performed on the symbolized candidate radar word set to obtain the reduced candidate radar word set. The reduced set of candidate radar words is parameterized and mapped to obtain the set of candidate operating modes.

12. The method according to claim 1, characterized in that, The decision-making process for the candidate work mode set is based on evidence theory, including: Obtain the set of candidate working modes The basic probability allocation function is derived from the expert review opinions. ;in The basic probability assignment function, The set of candidate working modes, It can be any one of the candidate working modes in the set of working modes; The confidence level of each working mode in the candidate working mode set is obtained by fusing the basic probability assignment function using the Dempster combination rule. ， in, The credibility level is stated above; The combination of operating modes with a confidence level greater than a preset confidence level threshold is determined as the operating mode of the multi-functional radar. ,in This refers to the operating mode of the multi-functional radar.

13. A multi-functional radar operating mode extraction device, characterized in that, include: The acquisition module is configured to acquire pulse sequence sample data of a specified type of multi-function radar from spaceborne passive detection data; The pulse sequence sample data includes N pulses, and the i-th pulse sequence sample data is... N is a positive integer greater than 2, and i is a positive integer greater than or equal to 1 and less than or equal to N; Let be the carrier frequency parameter of the i-th pulse sequence sample data. Let be the pulse width parameter of the i-th pulse sequence sample data. Let be the repetition interval parameter of the i-th pulse sequence sample data. and , Let i be the arrival time parameter of the i-th pulse sequence sample data; The clustering module is configured to cluster the carrier frequency parameter, the repetition interval parameter, and the pulse width parameter respectively, and perform multi-parameter joint symbolization and parameterization processing on the pulse sequence sample data based on the clustering results to obtain the symbolized pulse sequence and the parameterized pulse sequence, and establish a mapping between the symbolized pulse sequence and the parameterized pulse sequence. The radar word extraction module is configured to use a multi-order symbol combination pattern chain to extract candidate symbol combination sequences from the symbolized pulse sequence, and to determine a candidate radar word set based on the candidate symbol combination sequences. The mapping module is configured to perform symbolization processing on the candidate radar word set again, and to perform parameterized mapping processing on the symbolized candidate radar word set to obtain a candidate working mode set. The operating mode extraction module is configured to use evidence theory methods to make decisions on the candidate operating mode set and extract the operating mode of the multi-functional radar.

14. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 12.

15. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 12.

Images