A multi-frame joint detection method based on a graph neural network link prediction task

Abstract

The application provides a multi-frame joint detection method based on a graph neural network link prediction task, utilizes prior knowledge of target physical movement, carries out correlation test on observation data passing through a first threshold to construct a point track correlation graph, and then designs a multi-feature link prediction network based on a graph neural network, wherein the multi-feature link prediction network fuses echo structure, Doppler, space-time coupling and other features of multi-frame observation, and completes a multi-frame detection task. The application converts a multi-frame detection problem into link prediction in a graph, integrally completes track search and track decision, can directly output a target track from multi-frame observation to simplify processing logic, reduces performance loss of track search, and is beneficial to more accurate track decision. Compared with a traditional single-frame and multi-frame detection method, the application can effectively suppress false point tracks and improve target detection performance.

Description

Technical Field

[0001] The present invention relates to the field of radar technology, and in particular to a multi-frame joint detection method based on a graph neural network link prediction task, which can be used for long-distance weak target detection. Background Art

[0002] Target detection aims to determine whether a given echo originates from a target based on its signal strength. Effective target detection is the foundation for subsequent operations such as target tracking and identification. Traditional radar signal processing processes separate detection and tracking into two separate steps. This sequential approach of detecting first and then tracking is particularly problematic when dealing with targets with low signal-to-noise ratios. To successfully detect targets with low signal-to-noise ratios, it is often necessary to lower the threshold. However, lowering the threshold can lead to a large number of false alarms, making them difficult for the system to handle. It also increases the likelihood of misassociating the target with false alarms, making it difficult to correctly initiating the target's track.

[0003] To address the shortcomings of traditional signal processing pipelines, multi-frame detection (MFD) is considered a practical method for detecting faint targets. It improves detection performance for targets with low signal-to-noise ratios by fusing echo data from multiple consecutive frames. However, traditional multi-frame detection methods have significant drawbacks in high-resolution detection scenarios with a large number of detection units. On the one hand, the large number of raw observations imposes a heavy computational burden on the system, while on the other hand, the observation uncertainty caused by excessive false points makes it difficult to correctly associate targets. An effective countermeasure is the search-first, start-later multi-frame detection method. This method uses multiple frames of observations that pass a low threshold to search for potential tracks using the target's physical motion constraints. Finally, it makes a fusion decision along the potential tracks based on information such as echo energy. However, this method performs track search and track decision sequentially. This serial approach limits the final multi-frame detection performance to the target's potential track search performance. Furthermore, the model-driven track search approach employed by these methods makes it difficult to use unified constraint parameters to effectively associate targets with different motion characteristics. This results in a high incidence of missed target associations during track search. In addition, this type of method does not fully utilize the echo information and the space-time coupling characteristics of inter-frame traces, which is not conducive to multi-frame joint detection. Summary of the Invention

[0004] To address the above-mentioned problems in the prior art, the present invention provides a multi-frame joint detection method based on a graph neural network link prediction task. The technical problem to be solved by the present invention is achieved through the following technical solutions:

[0005] A multi-frame joint detection method based on graph neural network link prediction task, including:

[0006] S100, acquiring multiple frames of raw echo data, and setting a first threshold for detecting the raw echo data to obtain multiple frames of observation data; the observation data includes measurement information of multiple points;

[0007] S200, for each trace in the observation data of the current frame, performing an association test on the trace with traces in the observation data of subsequent frames of the current frame using a maximum velocity constraint; connecting traces that pass the association test with edges, and using the traces as nodes to obtain a trace association graph;

[0008] S300, using NFEN to perform feature mining on each node in the point trace association graph to obtain multidimensional extraction features, thereby obtaining multidimensional fusion features of each node, and forming a point trace extraction feature matrix with the multidimensional fusion features of all nodes;

[0009] S400, using STEF-GAN to mine multi-frame spatial-temporal coupling features in the point trace association graph, and during the mining process, using the point trace extraction feature matrix to perform message passing to obtain the final feature of each node;

[0010] S500, for any two nodes with a common edge, use OAJN to fuse the final features of the two nodes to calculate the classification score of the common edge between the two nodes, and remove the common edges that do not meet the classification score of the common edge, and retain the common edges between the targets.

[0011] Beneficial effects:

[0012] The present invention provides a multi-frame joint detection method GLP-MFD based on the graph neural network link prediction task. It uses the prior knowledge of the physical movement of the target to perform association tests on the observation data that passes the first threshold to construct a point track association graph. Then, a multi-feature link prediction network MFLPN designed based on the graph neural network is designed. The MFLPN integrates the echo structure, Doppler, space-time coupling and other features of single-frame observations to complete the multi-frame detection task. The present invention transforms the multi-frame detection problem into link prediction in the graph, and completes the track search and track judgment in an integrated manner. It can simplify the processing logic by directly outputting the target track from the multi-frame observation, which is conducive to more accurate track judgment while reducing the loss of track search performance. Compared with traditional single-frame and multi-frame detection methods, the present invention can improve the detection performance of the target while effectively suppressing false points.

[0013] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a flow chart of a multi-frame joint detection method based on a graph neural network link prediction task provided by the present invention;

[0015] Figure 2 The present invention provides Figure 1 Schematic diagram of the curve showing the relationship between detection probability and sliding window number when the signal-to-noise ratio is 9dB;

[0016] Figure 3 It is a curve diagram of the relationship between the detection probability and the number of sliding windows when the signal-to-noise ratio is 12 dB provided by the present invention. DETAILED DESCRIPTION

[0017] The present invention will be further described in detail below with reference to specific examples, but the embodiments of the present invention are not limited thereto.

[0018] like Figure 1 As shown, the present invention provides a multi-frame joint detection method GLP-MFD based on graph neural network link prediction task, including:

[0019] S100, acquiring multiple frames of raw echo data, and setting a first threshold for the raw echo data for detection to obtain multiple frames of observation data; the observation data includes measurement information of multiple points; the first threshold can be set in advance.

[0020] In a complex Gaussian white noise scenario containing a target, a coherent radar system can improve the signal-to-noise ratio of the target echo through pulse compression and non-coherent integration, while also obtaining Doppler measurement information. After radar target detection, the measurement information of the kth point in the nth frame can be expressed as follows:

[0021]

[0022] Where, t k,n Indicates the moment when the point is detected; r k,n Indicates the measured radial distance; θ k,n Indicates the measured azimuth; v k,n Indicates that there may be ambiguity in the measured radial velocity; d k,n Indicates the Doppler channel where the point is located; s k,n Indicates the signal-to-noise ratio of the trace; i k,n Indicates the time sequence code of the trace; a k,n The range Doppler diagram of the point trace; superscript Represents a transpose operation;

[0023] The total number of D in the nth frame n The traces are represented as:

[0024]

[0025] S200, for each trace in the observation data of the current frame, performing an association test on the trace with traces in the observation data of subsequent frames of the current frame using a maximum velocity constraint; connecting traces that pass the association test with edges, and using the traces as nodes to obtain a trace association graph;

[0026] There is a certain connection between the spatial and temporal characteristics of the target's multi-frame data between frames. Prior knowledge of the target's physical motion can be used to perform association tests on multi-frame observations that pass a low threshold. The point traces are used as nodes, and the successfully associated nodes are connected with edges to form an undirected graph, which the present invention calls a point trace association graph. The correct target track exists in the point trace association graph. The present invention needs to complete the integrated track search and track judgment process from the point trace association graph. The present invention uses the maximum speed constraint to construct the point trace association graph.

[0027] For two traces (t1, r1, θ1, v1) and (t2, r2, θ2, v2) in different frames, where t1 <t2,最大速度约束用公式表示为:

[0028]

[0029] In the formula, the two traces are represented as (t1, r1, θ1, v1) and (t2, r2, θ2, v2), v max Indicates the maximum speed of the target of interest, t1 <t2,r1,θ1,v1分别表示当前帧中点迹1的量测径向距离、量测方位角、可能存在模糊的量测径向速度;r2,θ2,v2分别表示后续帧中点迹2的量测径向距离、量测方位角、可能存在模糊的量测径向速度。

[0030] The task of multi-frame target detection is to observe Determine whether there is a target. In , two traces that satisfy the maximum velocity constraint are associated, where L represents the number of frames processed jointly.

[0031] S300, using NFEN to perform feature mining on each node in the point trace association graph to obtain multidimensional extraction features, thereby obtaining multidimensional fusion features of each node, and forming a point trace extraction feature matrix with the multidimensional fusion features of all nodes;

[0032] The NFEN designed by the present invention is effective for k,n ,s k,n ,i k,n ,a k,nThe feature mining is performed to enable the entire network to better learn the mapping from multi-dimensional attribute information to the track-derived target confidence. It includes three fully connected neural networks and one convolutional neural network. The fully connected layer is usually used for feature extraction and feature dimension transformation in NFEN. The convolutional structure not only reduces the number of parameters through parameter sharing, but also improves the NFEN's ability to learn local similarity information, thereby more effectively mining point trace features. This invention makes Represents the feature vector describing the point Doppler channel after feature mining, Represents the feature vector describing the signal-to-noise ratio of the trace after feature mining, Represents the feature vector that describes the time series information of the trace after feature mining. Represents the feature vector that describes the structural information of the trace signal after feature mining. k,n Represents the multidimensional fusion features that integrate different attribute information.

[0033] In a specific embodiment of the present invention, S300 includes:

[0034] S310, using NFEN to perform feature mining on each node in the point trace association graph to obtain multi-dimensional extracted features, the multi-dimensional extracted features including: a feature vector describing the point trace Doppler channel Eigenvector describing the signal-to-noise ratio of a trace Feature vector describing the timing information of the trace and the eigenvector describing the structural information of the trace signal

[0035] In a specific embodiment of the present invention, S310 includes:

[0036] S311, using the three fully connected neural networks of the NFEN module to calculate the d of each node in the point trace association graph k,n 、s k,n 、i k,n Perform feature mining to obtain the feature vector describing the point Doppler channel Eigenvector describing the signal-to-noise ratio of a trace and the feature vector describing the time series information of the trace in,

[0037]

[0038] Where, Represents the operations between layers in the NFEN module, Represents function composite operation, N d ,N s ,N i represents the number of fully connected network layers, Represents the learnable parameters of each layer of the network, φ D Represents the nonlinear activation function in the NFEN module, which can enhance the NFEN's ability to learn nonlinear mapping.

[0039] S312, using the convolutional neural network of NFEN to calculate the a of each node in the point trace association graph k,n Perform feature mining to obtain the feature vector that describes the structure information of the trace signal Expressed as:

[0040]

[0041] Where N p Indicates the number of convolution layers, * indicates the convolution operation, Represents the convolution kernel used in each layer of convolution operation.

[0042] S320 , for each node, fusing the multidimensional extracted features of the node to obtain a multidimensional fused feature, and forming a point trace extraction feature matrix with the multidimensional fused features of all nodes.

[0043] In a specific embodiment of the present invention, S320 includes:

[0044] S321, will Splicing is performed to obtain the multi-dimensional fusion feature f of each node k,n , expressed as:

[0045]

[0046] S322, multi-dimensional fusion features f of all nodes k,n Rewrite and form the trace extraction feature matrix f.

[0047] In a specific embodiment of the present invention, S322 includes:

[0048] S3221, rewrite the multidimensional fusion feature of each point trace according to the measurement time sequence of the point traces in multiple frames to obtain the rewritten multidimensional fusion feature, which is expressed as:

[0049]

[0050] Where L represents the total number of frames of observation data, l represents the lth frame, and there are D l A dot trace, Represents the rewritten multi-dimensional fusion features of the lth frame;

[0051] S3222, rewrite the multi-dimensional fusion features to form a point trace extraction feature matrix f.

[0052] S400, using STEF-GAN to mine multi-frame spatial-temporal coupling features in the point trace association graph, and during the mining process, using the point trace extraction feature matrix to perform message passing to obtain the final feature of each node;

[0053] In a specific embodiment of the present invention, S400 includes:

[0054] S410, using the trace extraction feature matrix f as the input node feature matrix x in STEF-GAN (0) , so that each layer of STEF-GAN transmits messages according to the message passing formula, and in the transmission process, the multi-frame spatial-temporal coupling features of the point trace association graph are mined to obtain the final features of each point trace; wherein, the message passing formula is expressed as:

[0055]

[0056] Where k represents the number of layers; Represents the features of node i in the kth layer, the input node feature matrix x of STEF-GAN (0) That is the point trace extraction feature matrix f output by the NFEN network; represents the set of neighbor nodes of node i; e j,i Represents the characteristics of the edge between node i and node j; A learnable matrix representing the features of nodes in the kth layer; Represents the learnable matrix of the k-th layer edge features, φ G Represents the nonlinear activation function in the GNN network, such as relu, sigmoid, softmax; Represents a fully connected layer; ⊕ represents an aggregation operation, such as addition, maximum value, and average, which can aggregate the neighbor node information of node i into itself; || represents feature merging; represents the attention coefficient of the kth layer; Represents the learnable parameters in the attention mechanism; represents the learnable matrix of the edge in the attention mechanism, and The difference is, Focus on learning the impact of edge features in the attention mechanism, It is used for feature enhancement of edges, focusing on their influence in aggregation operations. It is this message passing mechanism that enables STEF-GAN to fuse multi-frame structural information and explore the complex motion patterns between nodes, thereby increasing the separability of targets and false alarms.

[0057] There is a coupling relationship between the measurement time, measurement distance, measurement angle, and measurement speed of the target multi-frame traces with certain physical motion laws, but this coupling relationship is usually more complicated. k,n,r k,n ,θ k,n ,v k,n Directly inputting the node features into the network will increase the difficulty of network training and the computational burden. Based on this consideration, the t-values ​​between multiple frame traces can be pre-constructed based on the target domain knowledge. k,n ,r k,n ,θ k,n ,v k,n After obtaining the feature input of nodes and edges, STEF-GAN mines multi-frame information through multi-layer message passing to obtain the final feature, which can be expressed as follows:

[0058]

[0059] Where G· represents the inter-layer operation in STEF-GAN; Indicates function composite operation; N G represents the number of STEF-GAN network layers; e represents the edge feature matrix input to the STEF-GAN network.

[0060] After thoroughly exploring the multidimensional attributes of track points and the spatial and temporal coupling characteristics between track points across multiple frames, the present invention utilizes OAJN to perform intelligent track judgment within the track correlation graph. Compared to traditional judgment methods that rely on energy accumulation thresholds, this approach learns the mapping from multidimensional measurement information to whether track correlations originate from the target, enabling better assessment of track quality.

[0061] S500, for any two nodes with a common edge, use OAJN to fuse the final features of the two nodes to calculate the classification score of the common edge between the two nodes, and remove the common edges that do not meet the classification score of the common edge, and retain the common edges between the targets.

[0062] In a specific embodiment of the present invention, S500 includes:

[0063] S510, for any two nodes that have a common edge, using OAJN to fuse the final features of the two nodes, and calculating the classification score of the common edge between the two nodes;

[0064] Considering the edge feature e j,i It reflects the space-time coupling relationship between the traces. In order to make full use of this feature, the present invention will j,i Input it into OAJN again. In addition, the Doppler channel difference of the target is an important representation of the target motion characteristics. For example, for a target moving at a constant speed, the Doppler channel difference between different frames is usually small. The present invention calculates the Doppler channel difference d between node j and node i. j,i =|d j-d i | is also input into OAJN to better learn the classification prediction of edge attributes. In order to facilitate the subsequent setting of thresholds to distinguish target tracks from false tracks, the present invention selects the softamx function as the nonlinear activation during decoding. The edges in the point trace association graph are divided into three categories: false alarm-false alarm edges, false alarm-target edges, and target-target edges. Their true label values ​​are represented as For any two nodes with edges output by STEF-GAN and The score of the common edge between any two nodes is expressed as:

[0065]

[0066] Where, and Represents any two nodes, Indicates the scores of the edges between two nodes predicted as false alarm-false alarm edge, false alarm-target edge, and target-target edge, respectively. is the learnable parameter in OAJN; φ M represents the nonlinear activation function in OAJN, d j,i represents the Doppler channel difference between node i and node j.

[0067] S520, classifying the common edges into false alarm-to-false alarm edges, false alarm-to-target edges, and target-to-target edges according to the classification scores of the common edges and a second threshold;

[0068] S530 , removing edges between false alarms and edges between false alarms and targets as inconsistent common edges, and retaining common edges between targets.

[0069] In summary, the present invention provides a multi-frame joint detection method GLP-MFD based on the graph neural network link prediction task. After constructing a point track association graph for multiple consecutive frames of observation data that pass a low threshold, it uses a multi-feature link prediction MFLPN network to perform integrated track search and judgment. The multi-feature link prediction network MFLPN includes three modules: a node feature embedding network NFEN, an attention mechanism graph neural network STEF-GAN with space-time coupling edge features, and a point track association judgment network OAJN. NFEN extracts deep features of point track echo structure, Doppler and other information; STEF-GAN fully mines the association features containing complex physical motion laws between target point tracks; OAJN completes target detection through link prediction in the point track association graph based on the mined features. The present invention transforms the multi-frame detection problem into link prediction in the graph, and completes track search and judgment in an integrated manner, so as to simplify the processing logic by being able to directly output the target track from multi-frame observations, reducing the loss of track search performance while facilitating more accurate track judgment. Compared with traditional single-frame and multi-frame detection methods, the present invention can effectively suppress false points and improve the detection performance of the target.

[0070] The advantages of the present invention are reflected below through experimental content and results.

[0071] In order to demonstrate the improvement in detection performance brought about by the integrated multi-frame detection using the proposed GLP-MFD method, the present invention considers conducting simulation experiments in 2D scenes, but the proposed method can be easily extended to 3D scenes. The simulated radar parameters are set as follows: the radar operates in the L band, the beam width is 2°, the distance resolution is 100m, and the radar system transmits 32 pulses at each wave position. As mentioned above, the first threshold determines the optimal detection performance that can be achieved by the proposed method. Therefore, in order to ensure that the radar system has better detection performance for low signal-to-noise ratio targets, while not introducing too many false points that make the system calculation burden too heavy, the present invention sets the first threshold according to the false alarm probability P fa,1 =10 -3 set up.

[0072] In order to highlight the superiority of the proposed GLP-MFD method, the present invention uses the traditional search-first-then-start MFD method NDP and a model-driven track initiation combined with incoherent accumulation MFD method MTS-NCI for comparison. The NDP method uses target position measurement to construct kinematic constraints to obtain potential tracks, and fuses the track energy through the incoherent accumulation method to compare with the second threshold for track judgment. The MTS-NCI method makes full use of Doppler information, and constructs motion constraints based on the coupling relationship between position measurement and Doppler measurement to obtain potential tracks. By introducing Doppler information, the separability of target tracks and false tracks is increased, and then the track energy is incoherently accumulated and compared with the second threshold for track judgment. In order to make a fair comparison, the above methods all use the same window length for processing, and the false alarm probability after processing is guaranteed to be P for all methods. fa,2 =10 -6 .

[0073] The relationship curves between the target detection probability and the number of sliding windows for the above methods with signal-to-noise ratios of 9dB and 12dB are shown in the figure below. Figure 2 、 Figure 3 As shown. Among them, P D,CA-CFAR Indicates that the unit average constant false alarm method is used in a single frame with a false alarm probability of 10 -6 The probability of target detection under the condition P D,UPB Represents the upper limit of the detection performance of the proposed method and the comparison method. It can be seen that as the number of sliding windows increases, the detection probability of the multi-frame detection method gradually increases. In most cases, the multi-frame detection method can achieve better target detection performance than the single-frame detection method. Since the NDP method does not use Doppler information, the separability of targets and false alarms is poor. At the same time, its elimination of false tracks mainly relies on energy accumulation, which makes its single sliding window performance weaker than the traditional unit average constant false alarm method at low signal-to-noise ratio. Under high signal-to-noise ratio conditions, its performance is higher than the unit average constant false alarm method. The MTS-NCI method introduces Doppler information, which has better separability for targets and false alarms, and its performance is better than the NDP method. However, weak targets will reduce the separability of targets and false alarms, and setting the false alarm probability P fa,1 =10 -3 This results in a high number of false alarms passing the first threshold, leading to a performance bottleneck. The proposed GLP-MFD method fuses the multidimensional information of target traces with the spatial-temporal coupling between frames through a multi-feature link prediction network, approaching an optimal target detector that integrates multidimensional extracted features and achieving the highest detection performance. Furthermore, its superiority over other methods is even more pronounced at low signal-to-noise ratios, demonstrating that the proposed GLP-MFD method can effectively suppress false traces while maintaining good detection performance for weak targets.

[0074] It is worth noting that the terms "first" and "second" in this disclosure are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, features defined as "first" or "second" may explicitly or implicitly include one or more of such features. In the description of this disclosure, "plurality" means two or more, unless otherwise specifically defined.

[0075] Although the present application is described herein with reference to various embodiments, those skilled in the art will be able to understand and implement other variations of the disclosed embodiments in practicing the claimed application by reviewing the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude a plurality.

[0076] The above is a further detailed description of the present invention in conjunction with specific preferred embodiments, and the specific implementation of the present invention should not be considered to be limited to these descriptions. For those skilled in the art of the present invention, without departing from the concept of the present invention, several simple deductions or substitutions can be made, which should be considered to fall within the scope of protection of the present invention.

Claims

1. A multi-frame joint detection method based on graph neural network link prediction task, characterized by: include: S100, acquiring multiple frames of raw echo data, and setting a first threshold for the raw echo data to perform detection, thereby obtaining multiple frames of observation data; The observation data includes measurement information of multiple points; S200, for each trace in the observation data of the current frame, performing an association test on the trace with traces in the observation data of subsequent frames of the current frame using a maximum velocity constraint; connecting traces that pass the association test with edges, and using the traces as nodes to obtain a trace association graph; S300, using NFEN to perform feature mining on each node in the point trace association graph to obtain multidimensional extraction features, thereby obtaining multidimensional fusion features of each node, and forming a point trace extraction feature matrix with the multidimensional fusion features of all nodes; S400, using STEF-GAN to mine multi-frame spatial-temporal coupling features in the point trace association graph, and during the mining process, using the point trace extraction feature matrix to perform message passing to obtain the final feature of each node; S500, for any two nodes with a common edge, use OAJN to fuse the final features of the two nodes to calculate the classification score of the common edge between the two nodes, and remove the common edges that do not meet the classification score of the common edge, and retain the common edges between the targets.

2. The multi-frame joint detection method based on graph neural network link prediction task according to claim 1 is characterized in that The measurement information of the k-th point trace in the n-th frame in S100 is expressed as: Where, t k,n Indicates the moment when the point is detected; r k,n Indicates the measured radial distance; θ k,n Indicates the measured azimuth; v k,n Indicates that there may be ambiguity in the measured radial velocity; d k,n Indicates the Doppler channel where the point is located; s k,n Indicates the signal-to-noise ratio of the trace; i k,n Indicates the time sequence code of the trace; a k,n The range Doppler diagram of the point trace; superscript Represents a transpose operation; The total number of D in the nth frame n The traces are represented as:

3. The multi-frame joint detection method based on graph neural network link prediction task according to claim 2 is characterized in that The maximum speed constraint in S200 is expressed as: where the two tracks are respectively expressed as (t1, r1, θ1, v1) and (t2, r2, θ2, v2), and v max represents the maximum moving speed of the target of interest, t1 < t2, and r1, θ1, v1 respectively represent the measured radial distance, measured azimuth angle, and possibly ambiguous measured radial speed of Track 1 in the current frame; r2, θ2, v2 respectively represent the measured radial distance, measured azimuth angle, and possibly ambiguous measured radial speed of Track 2 in the subsequent frame.

4. The multi-frame joint detection method based on graph neural network link prediction task according to claim 3 is characterized in that S300 includes: S310, using NFEN mining to obtain multidimensional extraction features of each node in the point trace association graph, the multidimensional extraction features including: a feature vector describing the point trace Doppler channel Eigenvector describing the signal-to-noise ratio of a point trace Feature vector describing the timing information of the trace and the eigenvector describing the structural information of the trace signal S320 , for each node, fusing the multidimensional extracted features of the node to obtain a multidimensional fused feature, and forming a point trace extraction feature matrix with the multidimensional fused features of all nodes.

5. The multi-frame joint detection method based on graph neural network link prediction task according to claim 4 is characterized in that The S310 includes: S311, using the three fully connected neural networks of the NFEN module to calculate the d of each node in the point trace association graph k,n 、s k,n 、i k,n Perform the transformation to obtain the characteristic vector describing the point Doppler channel Eigenvector describing the signal-to-noise ratio of a point trace and the feature vector describing the time series information of the trace in, Where D represents the inter-layer operation in the NFEN module, Represents function composite operation, N d ,N s ,N i represents the number of fully connected network layers, Represents the learnable parameters of each layer of the network, φ D Represents the nonlinear activation function in the NFEN module; S312, using the convolutional neural network of NFEN to calculate the a of each node in the point trace association graph k,n Perform the transformation to obtain the feature vector describing the structure information of the trace signal Expressed as: Where N a Indicates the number of convolution layers, * indicates the convolution operation, Represents the convolution kernel used in each layer of convolution operation.

6. The multi-frame joint detection method based on graph neural network link prediction task according to claim 5 is characterized in that The S320 includes: S321, will Splicing is performed to obtain the multi-dimensional fusion feature f of each node k,n , expressed as: S322, multi-dimensional fusion features f of all nodes k,n Rewrite and form the trace extraction feature matrix f.

7. The multi-frame joint detection method based on graph neural network link prediction task according to claim 6 is characterized in that The S322 includes: S3221, rewrite the multidimensional fusion feature of each point trace according to the measurement time sequence of the point traces in multiple frames to obtain the rewritten multidimensional fusion feature, which is expressed as: Where L represents the total number of frames of observation data, l represents the lth frame, and there are D l A dot trace, Represents the rewritten multi-dimensional fusion features of the lth frame; S3222, rewrite the multi-dimensional fusion features to form a point trace extraction feature matrix f.

8. The multi-frame joint detection method based on graph neural network link prediction task according to claim 7 is characterized in that S400 includes: S410, using the trace extraction feature matrix f as the input node feature matrix x in STEF-GAN (0) , so that each layer of STEF-GAN transmits messages according to the message passing formula, and in the transmission process, the multi-frame spatial-temporal coupling features of the point trace association graph are mined to obtain the final features of each point trace; wherein, the message passing formula is expressed as: Where k represents the number of layers; Represents the features of node i in the kth layer, the input node feature matrix x of STEF-GAN (0) That is the point trace extraction feature matrix f output by the NFEN network; represents the set of neighbor nodes of node i; e j,i Represents the characteristics of the edge between node i and node j; A learnable matrix representing the features of nodes in the kth layer; Represents the learnable matrix of the k-th layer edge features, φ G Represents the nonlinear activation function in the GNN network; represents a fully connected layer; ⊕ represents an aggregation operation, and || represents feature merging; represents the attention coefficient of the kth layer; Represents the learnable parameters in the attention mechanism; A learnable matrix representing the edges in the attention mechanism; The final feature is expressed as: Where G · Represents the inter-layer operation in STEF-GAN; Indicates function composite operation; N G represents the number of STEF-GAN network layers; e represents the edge feature matrix input to the STEF-GAN network.

9. The multi-frame joint detection method based on graph neural network link prediction task according to claim 1 is characterized in that S500 includes: S510, for any two nodes that have a common edge, using OAJN to fuse the final features of the two nodes, and using the fused final features to calculate the score of the common edge between the two nodes; S520, classifying the common edges into false alarm-to-false alarm edges, false alarm-to-target edges, and target-to-target edges according to the classification scores of the common edges and a second threshold; S530: The edges between false alarms and the edges between false alarms and the target are regarded as inconsistent common edges, and the common edges between the target and the target are removed to obtain the retained common edges.

10. The multi-frame joint detection method based on graph neural network link prediction task according to claim 9 is characterized in that The classification score of the common edge between any two nodes is expressed as: Where, and Represents any two nodes, Indicates the scores of the edges between two nodes predicted as false alarm-false alarm edge, false alarm-target edge, and target-target edge, respectively. is the learnable parameter in OAJN; φ M represents the nonlinear activation function in OAJN, d j,i Represents the distance between node i and node j.

Images