Multifunctional radar signal sorting method and system based on hypergraph

Abstract

The invention discloses a multifunctional radar signal sorting method and system based on a hypergraph, and the method comprises the following steps: S1, eliminating interference pulses in an RIPS obtained through the interception of a radar reconnaissance receiver; s2, clustering the residual pulse sequences by using the proposed screening rule and FCM to obtain clustering labels of part of the pulse sequences; s3, constructing a hypergraph according to the characteristic parameters of the residual pulse sequences and the data potential energy values of the residual pulse sequences; and S4, learning the constructed hypergraph by using the clustering labels of the obtained partial pulse sequences to obtain a final radar signal sorting result. According to the method, the multifunctional radar signals can be sorted under the condition of no marked samples, the problem of batch increase can be effectively relieved, and good sorting performance is achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of radar reconnaissance, and in particular relates to a multi-function radar (MFRs, Multi-Function Radars) signal sorting method and system based on a hypergraph. The present invention can be used for electronic reconnaissance. Background Art

[0002] Radar signal sorting is a key part of radar reconnaissance technology, and its performance is crucial to subsequent processing such as interference, positioning and tracking. Usually, the input data of signal sorting is the pulse description words (PDW) of radar interleaved pulse sequences (RIPS), which mainly contain signal parameters such as direction of arrival (DOA), time of arrival (TOA), pulse width (PW), radio frequency (RF), etc. However, in a complex electromagnetic environment, the RIPS intercepted by the reconnaissance receiver will be affected by different degrees of noise and interference signals, and the parameters in the PDW obtained after signal detection and parameter estimation will inevitably have certain estimation errors. When these estimation errors are large, it is easy to cause some interference pulses to be contained in the RIPS, which seriously affects the radar signal sorting performance. In addition, with the continuous development of new phased array radar and other technologies, MFRs will automatically change their working modes according to the battlefield environment. The pulse parameters vary widely, and the correlation of pulse sequences formed under different working modes is low. Conventional sorting methods are prone to misjudge the pulse sequences of the same MFR in different working modes as pulse sequences of multiple radars, resulting in the problem of "increased batches". Therefore, in a complex electromagnetic environment with interference pulses and measurement errors, it is urgent to study how to effectively alleviate the sorting method of "increased batches" of MFRs signals.

[0003] Existing radar signal sorting methods can be divided into two categories: sorting methods based on pulse repetition interval (PRI) and sorting methods based on machine learning (ML). Sorting methods based on PRI mainly include cumulative difference histogram, sequence difference histogram and PRI transformation method. This type of method obtains the secondary parameter PRI by calculating TOA and uses PRI as the main characteristic parameter for sorting. This type of method has a relatively good sorting effect for radar signals with conventional pulse modulation. However, when faced with radar signals with various pulse modulation types generated by new MFRs, it exposes the shortcomings of large calculation amount and obvious decline in sorting performance. With the development of ML technology, researchers have begun to study radar signal sorting based on machine learning. At present, radar emitter intelligent sorting algorithms based on machine learning mainly include supervised models and unsupervised clustering models. Radar signal sorting based on supervised models has good sorting performance, but this type of method faces great challenges in training sample acquisition and algorithm real-time performance, especially on the battlefield, it is difficult to obtain a large number of labeled samples to train the model. In contrast, radar signal sorting based on unsupervised clustering models does not require labeled data, and does not take a lot of time to train the model. The specified hyperparameters are fewer than those of supervised sorting methods, which is more in line with actual needs. Although clustering methods sort radar signals from the perspective of unsupervised learning, they only have good sorting performance in simple scenarios such as different types of conventional radars and non-overlapping signal parameters. For MFRs signal sorting in multiple working modes, especially under non-ideal conditions such as interference pulses and measurement errors, the sorting performance of this type of method drops rapidly.

[0004] In recent years, researchers have begun to study the "increased batches" generated by MFRs signal sorting. Some researchers use the K nearest neighbor method to construct a graph structure and study the deinterlacing of MFRs signals from the perspective of residual graph convolutional networks. Although this approach can achieve good sorting performance, the method still requires a small amount of labeled sample data. In addition, some scholars have introduced complex networks into the field of radar signal sorting, providing a new idea for studying how to alleviate the "increased batches" of MFRs signal sorting. Some scholars have proposed a MFRs signal sorting method based on complex networks, using a sliding window limited penetration visibility graph to construct a complex network with RIPS, and using a label propagation algorithm and a density peak clustering algorithm to detect the community structure, thereby realizing MFRs signal sorting. However, this method first constructs a complex network for a single feature parameter, and then simply merges each complex network to obtain a total complex network. This network construction method does not consider the correlation between feature parameters when constructing the network. Based on the deficiencies of existing research, the present invention proposes to use a hypergraph method to construct a network. The construction of traditional graphs only considers the correlation of a single feature in the data, while the hypergraph exhibits natural high-order correlation in structure, and the constructed hypergraph contains the correlation of multiple features between the data. Therefore, compared with the graph structure obtained by using complex network methods, when constructing a hypergraph from the PDW sequence, the correlation between the parameters in the PDW is considered, which is more conducive to radar signal sorting.

[0005] Based on the above research status, the present invention proposes a multifunctional radar signal sorting method and system based on hypergraph. Summary of the invention

[0006] Aiming at the problem of "increased batches" that easily occurs during MFRs signal sorting, and the inevitable existence of interference pulses and parameter estimation errors in RIPS intercepted in complex electromagnetic environments, which lead to a decrease in sorting performance, the present invention proposes a multi-function radar signal sorting method and system based on hypergraph.

[0007] In order to achieve the above purpose, the present invention adopts the following technical solutions:

[0008] A multifunctional radar signal sorting method based on a hypergraph comprises the following steps:

[0009] S1. For the RIPS intercepted by the radar reconnaissance receiver, remove the interference pulse in the RIPS;

[0010] S2, clustering the remaining pulse sequences by using the screening rule and fuzzy C-means (FCM) to obtain cluster labels for the remaining pulse sequences;

[0011] S3, constructing a hypergraph using the characteristic parameters of the remaining pulse sequences and their data potential energy values;

[0012] S4. Use the clustering labels of the remaining pulse sequences to learn the constructed hypergraph to obtain the final radar signal sorting result.

[0013] Furthermore, in step S1, the interference pulse is removed using the data field method, and the specific steps are as follows:

[0014] Step S11, performing maximum and minimum normalization on the three characteristic parameters of RF, PW and DOA in the PDW data:

[0015]

[0016] Among them, x d represents the d-dimensional feature parameter vector, represents the dth feature of the ith pulse;

[0017] Step S12, calculating the data potential energy value of each PDW data:

[0018]

[0019] Among them, pdw m represents the mth PDW, d(pdw i ,pdw j ) means pdw i With pdw j The Euclidean distance between them, σ is the radiation factor, N is the number of pulses, F(pdw m ) means pdw m Data potential value;

[0020] Step S13, calculating the detection threshold of the data potential energy value, considering the data points whose data potential energy value is much smaller than the average data potential energy value as interference pulses, and removing them, which is specifically expressed as:

[0021]

[0022] in, is the detection threshold, F(pdw jl ) is the interference pulse pdw jl The data potential energy value. ξ is the detection coefficient set according to the electromagnetic environment, mean represents the average value, represents the sum of the data potential energy values ​​of all pulses, N is the number of pulses, F(pdw i ) is the data potential energy value of the i-th PDW.

[0023] Furthermore, in step S2, on the one hand, the proposed screening rules are used to obtain local core point data for the remaining PDW data, and on the other hand, FCM clustering is performed on the remaining PDW data to obtain the membership value of each pulse. Then, the local core points are traversed to obtain the cluster labels of local core points greater than the set membership value, and the cluster labels of these local core points are used as known information for subsequent learning, as follows:

[0024] Step S21, filtering the remaining PDW data according to the proposed filtering rule to obtain local core points; the filtering rule is expressed as:

[0025]

[0026] Among them, μ(δ) represents the relative distance mean of all data, μ(ρ) represents the local density mean of all data, σ(ρ) represents the local density standard deviation of all data, μ(RNN) represents the reverse nearest neighbor number mean of all data, μ and σ represent the mean and standard deviation respectively, local_core_pdw i represents i local core points, δ i represents the relative distance of the i-th data, δ represents the relative distance of all data, and ρ i represents the local density of the i-th data, ρ represents the local density of all data, |RNN i | represents the number of reverse nearest neighbors of the i-th data, and |RNN| represents the number of reverse nearest neighbors of all data;

[0027] Step S22, estimate the number of MFRs P, and determine P based on the inflection point of the cluster sum of squares (CSS) and the maximum value of the silhouette coefficient (SC); CSS and SC are expressed as:

[0028]

[0029] Where c represents the number of cluster centers, n represents the number of remaining pulses, and d(C i ,pdw j ) represents the i-th cluster center C i With the jth PDW data pdw j The distance between

[0030]

[0031] Among them, a(pdw i ) is pdw i The average distance to other PDW data points in the same cluster, b(pdw i ) is pdwi To its nearest neighbor cluster (i.e., not including pdw i The average distance of all PDW data points in the cluster where the PDW data points are located;

[0032] Step S23, performing FCM clustering on the remaining PDW data according to the MFR quantity P obtained in step S22, and obtaining the membership value u of each PDW data ij , traverse the membership values ​​of the local core points obtained in step S21, when a certain membership value of a local core point is greater than a set value, extract the clustering label of the local core point; calculate the membership value u ij It is expressed as:

[0033]

[0034] Among them, u ij The jth PDW data belongs to the i-th cluster center C i The membership value of is , m is the fuzzy index, which is usually an integer greater than 1.

[0035] Furthermore, in step S3, the hyperedge of each node is constructed for the remaining PDW data using the K-nearest neighbor (KNN) method, thereby obtaining a hypergraph; the hypergraph structure is represented by an association matrix H, which is represented as:

[0036]

[0037] Among them, v H represents a node in the hypergraph, e H Represents a hyperedge in a hypergraph.

[0038] Furthermore, in step S4, the steps of learning the constructed hypergraph are as follows:

[0039] Step S41, constructing a node classification function F; taking two types of label prediction as an example, the node classification function F(v H ) is expressed as:

[0040]

[0041] Among them, S and Indicates that the entire hypergraph is divided into two parts, one part is represented by S and the remaining part is represented by express;

[0042] Step S42, construct a regularizer Ω(F) on the hypergraph, through F(v H ) represents the regularizer Ω(F) as:

[0043]

[0044] Where c(S) represents the hypergraph partition, Δ is the normalized Laplacian matrix, and E H represents the set of hyperedges, e H represents a hyperedge, w(e H ) represents the hyperedge e H The weight of δ(e H ) means e H The super edge degree, u H and v H represents two nodes on the hypergraph, F(u H ) represents u H The label of d(u H ) represents u H The node degree, F(v H ) indicates v H The label of d(v H ) indicates v H The node degree of

[0045] Step S43, the objective function of the hypergraph learning model can be divided into two parts. The first part is the empirical loss R of the learning label. emp (F), such as the least squares loss. The second part is the regularizer Ω(F) on the hypergraph, which represents the smoothness of the label distribution on the constructed hypergraph structure; through linear combination, the minimization objective function of hypergraph learning is expressed as:

[0046]

[0047] in, represents the minimization objective function of hypergraph learning, and λ is the trade-off coefficient;

[0048] S44, the hypergraph learning minimization objective function in step S43 is further derived and expressed as:

[0049]

[0050] Among them, Y represents the initial label vector, for the labeled node v H , if the node belongs to the jth class, then Y(v H ) is 1, and the other positions are 0; for the unmarked node u H ,Y(u H ) are all set to 0.5, indicating that this part of unlabeled nodes has no prior category information;

[0051] When the hypergraph learning objective function reaches the minimum and converges, the node classification function is expressed as:

[0052]

[0053] Among them, Dv represents the node degree matrix, D e represents the hyperedge degree matrix, μ generally takes values ​​between 0 and 1;

[0054] In step S45, when the hypergraph learning objective function reaches the minimum and converges, the label message transmission on the hypergraph reaches a balanced state, each node has corresponding label information, and the radar signal sorting can be completed by classifying the nodes with different labels.

[0055] The present invention also discloses a multifunctional radar signal sorting system based on a hypergraph, which is used to execute the above method, and specifically comprises the following modules:

[0056] Interference pulse removal module: removes interference pulses in the intercepted radar interleaved pulse sequence RIPS;

[0057] The partial local core point label extraction module uses screening rules and fuzzy C-means FCM clustering to obtain the cluster labels of the remaining pulse sequences;

[0058] Hypergraph construction module: constructs a hypergraph with the characteristic parameters of the remaining pulse sequences and their data potential energy values;

[0059] Hypergraph learning module: Use the clustering labels of the remaining pulse sequences to learn the constructed hypergraph;

[0060] Sorting module: When the hypergraph learning in the hypergraph learning module reaches convergence, the label information of each node can be obtained, and the nodes can be classified to obtain the radar signal sorting results.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] 1. Taking into account the correlation between feature parameters, the present invention proposes to use a hypergraph method to construct a network for the multi-dimensional feature parameters of a pulse sequence. The obtained hypergraph structure reflects the correlation between the feature parameters of data samples, ensures the accuracy of hypergraph learning, and thus improves the sorting performance.

[0063] 2. The present invention uses the data field method to remove the interference pulses in RIPS, and uses the obtained data potential energy value of each pulse as a characteristic parameter, combined with the existing PDW parameters for use in subsequent stages. The data potential energy value can characterize the local density of each pulse data, and to a certain extent reflects the spatial concentration of PDW data. Therefore, adding the data potential energy value to the PDW parameters can effectively assist in MFRs signal sorting.

[0064] 3. Taking into account that the actually intercepted pulse sequence contains less known information, the present invention proposes a new sample screening rule, which screens the remaining pulse sequence after removing the interference pulses to obtain local core points, and combines FCM clustering to obtain clustering labels of some local core points, thereby automatically obtaining the label data required for subsequent hypergraph learning.

[0065] 4. The present invention can sort pulse sequences of different working modes of the same MFR as the same radar radiation source, effectively alleviating the problem of "increased batches" in sorting. The sorting performance of the present invention is superior to the existing complex network-based methods and unsupervised clustering-based methods, and has stronger robustness under non-ideal conditions.

[0066] In summary, in view of the problem of "increased batches" that is prone to occur during MFRs signal sorting, and the inevitable interference pulses and parameter estimation errors in the RIPS intercepted in a complex electromagnetic environment, which lead to a decrease in sorting performance, the present invention proposes a multifunctional radar signal sorting method and system based on a hypergraph, introduces the hypergraph method into the field of radar signal sorting, uses the data field method to eliminate interference pulses in RIPS, and proposes a screening rule to screen local core points, and combines FCM clustering to extract clustering labels of some local core points, so that MFRs signals can be sorted without prior information in the actual battlefield. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 A flowchart of a multifunctional radar signal sorting method based on a hypergraph according to a preferred embodiment of the present invention;

[0068] Figure 2 A flow chart of eliminating interference pulses in RIPS in a preferred embodiment of the present invention;

[0069] Figure 3 A flowchart of extracting cluster labels of some local core points in a preferred embodiment of the present invention;

[0070] Figure 4 A schematic diagram of hyperedge segmentation in a preferred embodiment of the present invention;

[0071] Figure 5 It is a two-dimensional visualized distribution diagram of original data containing interference pulses and a two-dimensional visualized distribution diagram of remaining data after removing interference pulses using the present invention;

[0072] Figure 6 The two-dimensional visualization distribution map of the local core points obtained by screening and the two-dimensional visualization distribution map of some local core points with clustering labels;

[0073] Figure 7 It is a two-dimensional visualized distribution diagram after the ablation experiment of the present invention and a two-dimensional sorting result diagram obtained by using the present invention;

[0074] Figure 8 Two-dimensional visualization of the distribution of the actual results using spectral clustering, First Integer Neighbor Clustering Hierarchy (FINCH) and complex network sorting methods;

[0075] Fig. 9 It is the sorting accuracy of the present invention under different parameter estimation error ratios and different leakage pulse rates;

[0076] Fig.10 It is a block diagram of a multifunctional radar signal sorting system based on a hypergraph according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0077] The following describes the implementation process of the present invention through preferred embodiments, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention.

[0078] See also Figure 1 This embodiment provides a multi-function radar signal sorting method based on a hypergraph. Taking four radars as an example, two multi-function radars are included. Radar1 and Radar3 are conventional radars, and Radar2 and Radar4 are multi-function radars, each of which has two working modes. X represents the intercepted PDW data, J represents the PDW data of the interference pulse, and C X (J) represents the PDW data of the remaining pulse, C XL (J) is the PDW data of the local core point, U is the membership matrix of the remaining PDW data, C X1 (J) represents the sorting result of Radar1, and so on, specifically including the following steps:

[0079] S1. For the RIPS intercepted by the radar reconnaissance receiver, the characteristic parameters of RIPS are first normalized to the maximum and minimum, and then the data field potential energy values ​​of all pulse sequences are calculated to determine whether the data potential energy values ​​of all pulse sequences are greater than the detection threshold. Pulses less than the detection threshold are interference pulses, and finally the interference pulses are eliminated. For specific steps, refer to Figure 2 .

[0080] Step S11: perform maximum and minimum normalization on the three characteristic parameters of RF, PW and DOA in the PDW data:

[0081]

[0082] Among them, x d represents the d-dimensional feature parameter vector, represents the dth feature of the ith pulse.

[0083] Step S12: Calculate the data potential energy value of each PDW data. The calculated data potential energy value can be expressed as:

[0084]

[0085] Among them, pdw m represents the mth PDW, d(pdw i ,pdw j ) means pdw i With pdw j The Euclidean distance between them, σ is the radiation factor, N is the number of pulses, F(pdw m ) means pdw m The data potential value.

[0086] Step S13, calculating the detection threshold of the data potential energy value, considering the data points whose data potential energy value is much smaller than the average data potential energy value as interference pulses, and eliminating them. The detection threshold can be expressed as;

[0087]

[0088] in, is the detection threshold, F(pdw jl ) is the interference pulse pdw jl The data potential energy value. ξ is the detection coefficient set according to the electromagnetic environment, mean represents the average value, represents the sum of the data potential energy values ​​of all pulses, N is the number of pulses, F(pdw i ) is the data potential energy value of the i-th PDW.

[0089] S2. After removing the interference pulses in step S1, on the one hand, the proposed screening rules are used to obtain the local core point data for the remaining PDW data, and on the other hand, the FCM clustering is performed on the remaining PDW data to obtain the membership value of each pulse. Then the local core points are traversed to obtain the clustering labels of the local core points that are greater than the set membership value, and the clustering labels of these local core points are used as known information for subsequent learning. For specific steps, refer to Figure 3 .

[0090] Step S21: Filter the remaining PDW data according to the proposed filtering rules to obtain local core points. The filtering rules can be expressed as:

[0091]

[0092] Among them, μ(δ) represents the relative distance mean of all data, μ(ρ) represents the local density mean of all data, σ(ρ) represents the local density standard deviation of all data, μ(|RNN|) represents the reverse nearest neighbor number mean of all data, μ and σ represent the mean and standard deviation respectively, local_core_pdw i represents i local core points, δ i represents the relative distance of the i-th data, δ represents the relative distance of all data, and ρ i represents the local density of the i-th data, ρ represents the local density of all data, |RNN i | represents the number of reverse nearest neighbors of the i-th data, and |RNN| represents the number of reverse nearest neighbors of all data;

[0093] Step S22, estimate the MFR quantity P, and determine P according to when CSS has an inflection point and SC reaches a maximum value. CSS and SC can be expressed as:

[0094]

[0095] Where c represents the number of cluster centers, n represents the number of remaining pulses, and d(C i ,pdw j ) represents the i-th cluster center C i With the jth PDW data pdw j The distance between

[0096]

[0097] Among them, a(pdw i ) is pdw i The average distance to other PDW data points in the same cluster, b(pdw i ) is pdw i To its nearest neighbor cluster (i.e., not including pdw i The average distance of all PDW data points in the cluster.

[0098] Step S23: Perform FCM clustering on the remaining PDW data according to the MFR quantity P obtained in step S22 to obtain the membership value u of each PDW data. ij , the membership value of the local core point obtained by traversal, when a certain membership value of a local core point is greater than 0.5, the clustering label of the local core point is extracted and the membership value u is calculated ij It can be expressed as:

[0099]

[0100] Among them, u ij The jth PDW data belongs to the i-th cluster center Ci The membership value of is , m is the fuzzy index, which is usually an integer greater than 1.

[0101] S3. Use the KNN method to construct the hyperedge of each node for the remaining PDW data, and then obtain a hypergraph. The hypergraph structure is usually represented by the association matrix H, which can be expressed as:

[0102]

[0103] Among them, v H represents a node in the hypergraph, e H Represents a hyperedge in a hypergraph.

[0104] S4. Use the clustering labels of the obtained partial pulse sequences to learn the constructed hypergraph to obtain the final radar signal sorting result. The specific steps are as follows:

[0105] Step S41: Construct node classification function F(v H ), taking two types of label prediction as an example, the node classification function F(v H ) can be expressed as;

[0106]

[0107] Among them, S and Indicates that the entire hypergraph is divided into two parts, one part is represented by S and the remaining part is represented by Indicates that specific reference Figure 4 .

[0108] Step S42: construct a regularizer Ω(F) on the hypergraph, and use F(v H )The regularizer Ω(F) can be expressed as;

[0109]

[0110] Where c(S) represents the hypergraph partition, Δ is the normalized Laplacian matrix, and E H represents the set of hyperedges, e H represents a hyperedge, w(e H ) represents the hyperedge e H The weight of δ(e H ) means e H The super edge degree, u H and v H represents two nodes on the hypergraph, F(u H ) represents u H The label of d(u H ) represents u H The node degree, F(v H ) indicates v H The label of d(vH ) indicates v H The node degree of

[0111] Step S43: The objective function of the general hypergraph learning model can be divided into two parts. The first part is the empirical loss R of the learning label. emp (F), such as the least squares loss. The second part is the regularizer Ω(F) on the hypergraph, which represents the smoothness of the label distribution on the constructed hypergraph structure. Through linear combination, the minimization objective function of hypergraph learning can be expressed as;

[0112]

[0113] in, represents the minimization objective function of hypergraph learning, and λ is the trade-off coefficient.

[0114] Step S44, further deriving the hypergraph learning minimization objective function in step S43 can be expressed as;

[0115]

[0116] Among them, Y represents the initial label vector, for the labeled node v H , if the node belongs to the jth class, then Y(v H ) is 1, and the other positions are 0. For an unlabeled node u H ,Y(u H ) are all set to 0.5, indicating that there is no prior category information for these unlabeled nodes.

[0117] When the hypergraph learning objective function reaches the minimum and converges, the node classification function can be expressed as;

[0118]

[0119] Among them, D v represents the node degree matrix, D e Represents the hyperedge degree matrix, μ generally takes a value between 0 and 1.

[0120] Step S45, when the hypergraph learning reaches convergence, the label message transmission on the hypergraph reaches a balanced state, each node has corresponding label information, and the radar signal sorting can be completed by classifying the nodes with different labels.

[0121] According to the above embodiment, the experiment was conducted: two multi-function radars and two conventional radar data sets were used as examples to verify the effectiveness of sorting. In combination with the real scene, MFR will switch different working modes in real time. Radiation source 1 and radiation source 3 only work in one mode, and radiation source 2 and radiation source 4 work in three modes respectively. PW, RF and PRI have multiple modulation types such as jitter, stagger and agility. In addition, in addition to testing the sorting effect of different numbers of interference pulses, the proposed algorithm uniformly sets the number of interference pulses to 50 when testing the sorting performance, and the total number of RIPS is 2450.

[0122] The present invention is used to remove the interference pulses in RIPS. The visualization results can be found in Figure 5 . Figure 5 In (a), the black five-pointed star represents an interference pulse, the small circle represents a normal pulse, PW represents pulse width normalization, and RF represents carrier frequency normalization. Similarly, in other figures, it can be seen that the distribution of interference data points is relatively sparse, while most other pulse sequences have data clustering and are relatively densely distributed. (b) shows the distribution of the remaining data points after eliminating the interference data points. It can be seen that after eliminating the data points with data potential energy values ​​less than the detection threshold value, there are very few interference data points in the remaining data. The interference pulses in RIPS are eliminated, and then the present invention is used to obtain the clustering labels of some PDW data for the remaining pulse data. For the visualization results of the local core points and the clustering labels of some local core points obtained by screening, please refer to Figure 6 shown. Figure 6 In (a), the inverted triangle represents the local core points obtained through screening. The distribution characteristics of the local core points obtained through screening are that the local core points of each part are relatively dense, and the distance between the local core points of adjacent parts is relatively far. (b) shows the distribution of some local core points with cluster labels. The four gray diamonds represent the four cluster center points, and the four different shapes of graphics represent the local core point cluster labels of the four types of data. Figure 7 The ablation experiment of each step of the present invention is shown, (a), (b) and (c) are respectively without data field elimination interference pulse, without screening rules and without hypergraph method, (d) is the sorting result obtained by using the present invention, Figure 7 In this example, predict-radar1 indicates the distribution of sorting results predicted for radar 1, and so on. The same is true for other figures. Figure 7 This shows the effectiveness and importance of each step of the method proposed in the present invention. Figure 8 The comparison results of the proposed method with spectral clustering, FINCH clustering and complex network are shown in Figure 2. To ensure fairness, no interference pulses were added to the three comparison methods. Specifically, Figure 8In the figure, (a) is the sorting result obtained by spectral clustering method, (b) is the sorting result obtained by FINCH clustering, (c) is the sorting result obtained by complex network method, and (d) is the real result distribution after removing interference pulses by data field method, where radar1 represents the data distribution of real radar 1, and so on. Interferencepulse represents the remaining interference pulses. It can be seen that there are very few interference pulses in (d). Figure 8 It shows that the method proposed in the present invention is better than the other three methods in sorting multi-function radar signals. Fig. 9 It shows the performance effect of the sorting of the method proposed in the present invention under different parameter estimation error ratios and different leakage pulse rates, where Proportion represents different error ratios or different leakage pulse rate proportions, Accuracy represents the sorting accuracy, Radar1 represents the sorting accuracy of radar 1 under different parameter estimation error ratios and different leakage pulse rates, and so on. It can be seen from (a) that when the estimation error is less than or equal to 2%, the sorting accuracy of each radar is above 86%, and the sorting accuracy of three radars is 96%. It can be seen from (b) that when the leakage pulse rate is less than or equal to 25%, the sorting accuracy of each radar is above 86%, indicating that the method proposed in the present invention is robust. Therefore, the multifunctional radar sorting method based on the hypergraph in this embodiment has good sorting performance for sorting signals containing MFRs, laying a foundation for further processing such as identification of radar signals.

[0123] like Fig.10 As shown, this embodiment discloses a multifunctional radar signal sorting system based on a hypergraph, which is used to execute the above method embodiment, and specifically includes the following modules:

[0124] Interference pulse removal module: For the intercepted radar interleaved pulse sequence RIPS, the interference pulses in RIPS are removed; specifically, the characteristic parameters of RIPS are first normalized to the maximum and minimum, and then the data field potential energy values ​​of all pulse sequences are calculated to determine whether the data potential energy values ​​of all pulse sequences are greater than the detection threshold. Pulses less than the detection threshold are interference pulses, and finally the interference pulses are removed;

[0125] Partial local core point label extraction module: The remaining pulse sequence is clustered using screening rules and fuzzy C-means FCM to obtain the clustering labels of the remaining pulse sequence; specifically, the proposed screening rules are used to screen the characteristic parameters of the remaining pulses to obtain local core points, and the FCM clustering algorithm is combined to extract the clustering labels of some local core points as known labels for the subsequent sorting stage;

[0126] Hypergraph construction module: construct a hypergraph with the characteristic parameters of the remaining pulse sequence and its data potential energy value; specifically, use the KNN method to perform hypergraph modeling on the characteristic parameters and data potential energy values ​​of the remaining pulses;

[0127] Hypergraph learning module: The cluster labels of the remaining pulse sequences are used to learn the constructed hypergraph. Specifically, the cluster labels of some local core points obtained in the partial local core point label extraction module are used as known information. The remaining unlabeled nodes are learned by first learning some labeled node data on the hypergraph.

[0128] Sorting module: When the objective function in the hypergraph learning module reaches the minimum and converges, the label information of each node can be obtained, and the radar signal sorting results can be obtained by classifying each node.

[0129] For other contents of this embodiment, please refer to the above method embodiment.

[0130] The above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A multifunctional radar signal sorting method based on a hypergraph, characterized in that: The specific steps include: S1. Eliminate the interference pulses in the intercepted radar interleaved pulse sequence RIPS; S2, clustering the remaining pulse sequences using screening rules and fuzzy C-means FCM to obtain cluster labels of the remaining pulse sequences; S3, constructing a hypergraph using the characteristic parameters of the remaining pulse sequences and their data potential energy values; S4. The clustering labels of the remaining pulse sequences are used to learn the constructed hypergraph to obtain the final radar signal sorting result.

2. The multifunctional radar signal sorting method based on hypergraph according to claim 1 is characterized in that: In step S1, the interference pulses in the RIPS are eliminated using the data field method, as follows: S11, perform maximum and minimum normalization on the three characteristic parameters of carrier frequency RF, pulse width PW and direction of arrival DOA in the pulse description word PDW data of RIPS: Among them, x d represents the d-dimensional feature parameter vector, represents the dth feature of the ith pulse; S12, calculating the data potential energy value of each PDW data, and the calculated data potential energy value is expressed as: Among them, pdw m represents the mth PDW, d(pdw i ,pdw j ) means pdw i With pdw j The Euclidean distance between them, σ is the radiation factor, N is the number of pulses, F(pdw m ) means pdw m Data potential value; S13, calculating the detection threshold of the data potential energy value, identifying the data points whose data potential energy value is less than the set average data potential energy value as interference pulses, and removing them, which is specifically expressed as: in, is the detection threshold, F(pdw jl ) is the interference pulse pdw jl The data potential energy value; ξ is the detection coefficient set according to the electromagnetic environment, mean represents the average value, represents the sum of the data potential energy values ​​of all pulses, N is the number of pulses, F(pdw i ) is the data potential energy value of the i-th PDW.

3. The multifunctional radar signal sorting method based on hypergraph according to claim 2 is characterized in that: Step S2 is specifically as follows: S21, filtering the remaining pulse sequences according to the filtering rule to obtain the local core points, wherein the filtering rule is expressed as: Among them, μ(δ) represents the relative distance mean of all data, μ(ρ) represents the local density mean of all data, σ(ρ) represents the local density standard deviation of all data, μ(RNN) represents the reverse nearest neighbor number mean of all data, μ and σ represent the mean and standard deviation respectively, local_core_pdw i represents i local core points, δ i represents the relative distance of the i-th data, δ represents the relative distance of all data, and ρ i represents the local density of the i-th data, ρ represents the local density of all data, |RNN i Represents the number of reverse nearest neighbors of the i-th data, and |RNN| represents the number of reverse nearest neighbors of all data; S22. Estimate the number of multi-function radar MFRs P. P is determined when the inflection point of the cluster sum of squares CSS appears and the contour coefficient SC reaches the maximum value. CSS and SC are expressed as: Where c represents the number of cluster centers, n represents the number of remaining pulses, and d(C i ,pdw j ) represents the i-th cluster center C i With the jth PDW data pdw j The distance between Among them, a(pdw i ) is pdw i The average distance to other PDW data points in the same cluster, b(pdw i ) is pdw i The average distance to all PDW data points in its nearest neighbor cluster; S23, perform FCM clustering on the remaining pulse sequences according to the MFR number P obtained in step S22 to obtain the membership value u of each pulse sequence ij , the membership value of the local core point obtained by traversal, when a certain membership value of a local core point is greater than the set value, the clustering label of the local core point is extracted and the membership value u is calculated ij It is expressed as: Among them, u ij The jth PDW data belongs to the i-th cluster center C i The membership value of is, and m is the fuzzy index.

4. The multifunctional radar signal sorting method based on hypergraph according to claim 3 is characterized in that: In step S3, the K nearest neighbor (KNN) method is used to construct the hyperedge of each node for the remaining pulse sequences to obtain a hypergraph.

5. The multifunctional radar signal sorting method based on hypergraph according to claim 4 is characterized in that: In step S3, the hypergraph structure is represented by an association matrix H, which is expressed as: Among them, v H represents a node in the hypergraph, e H Represents a hyperedge in a hypergraph.

6. The multifunctional radar signal sorting method based on hypergraph according to claim 5 is characterized in that: Step S4 is specifically as follows: S41, construct node classification function F(v H ), taking two types of label prediction as an example, the node classification function F(v H ) is expressed as: Among them, S and S indicate that the entire hypergraph is divided into two parts, one part is represented by S and the remaining part is represented by S; S42, construct a regularizer Ω(F) on the hypergraph, through F(v H ) represents the regularizer Ω(F) as: Where c(S) represents the hypergraph partition, Δ is the normalized Laplacian matrix, and E H represents the set of hyperedges, e H represents a hyperedge, w(e H ) represents the hyperedge e H The weight of δ(e H ) means e H The super edge degree, u H and v H represents two nodes on the hypergraph, F(u H ) represents u H The label of d(u H ) represents u H The node degree, F(v H ) indicates v H The label of d(v H ) indicates v H The node degree of S43. The objective function in the hypergraph learning model is divided into two parts. The first part is the empirical loss R of the learning label. emp (F), the second part is the regularizer Ω(F) on the hypergraph, which represents the smoothness of the label distribution on the constructed hypergraph structure; through linear combination, the minimization objective function of hypergraph learning is expressed as: in, represents the minimization objective function of hypergraph learning, and λ is the trade-off coefficient; S44, the hypergraph learning minimization objective function in step S43 is further derived and expressed as: Among them, Y represents the initial label vector, for the labeled node v H , if the node belongs to the jth class, then Y(v H ) is 1, and the other positions are 0; for the unmarked node u H ,Y(u H ) are all set to 0.5, indicating that this part of unlabeled nodes has no prior category information; When the hypergraph learning objective function reaches the minimum and converges, the node classification function is expressed as: Among them, D v represents the node degree matrix, D e represents the hyperedge degree matrix, μ generally takes values ​​between 0 and 1; S45. When the hypergraph learning reaches convergence, the label message transmission on the hypergraph reaches a balanced state, each node has corresponding label information, and the radar signal sorting is completed by classifying the nodes with different labels.

7. A multifunctional radar signal sorting system based on a hypergraph, used to execute the method according to any one of claims 1 to 6, characterized in that Includes the following modules: Interference pulse removal module: removes interference pulses in the intercepted radar interleaved pulse sequence RIPS; Partial local core point label extraction module: The remaining pulse sequence is clustered using screening rules and fuzzy C-means FCM to obtain the cluster label of the remaining pulse sequence; Hypergraph construction module: constructs a hypergraph with the characteristic parameters of the remaining pulse sequences and their data potential energy values; Hypergraph learning module: Use the clustering labels of the remaining pulse sequences to learn the constructed hypergraph; Sorting module: When the hypergraph learning in the hypergraph learning module reaches convergence, the label information of each node is obtained, and each node is classified to obtain the radar signal sorting result.

Images