Electromagnetic target intelligent identification method and system based on spatio-temporal joint perception

Abstract

The invention provides an electromagnetic target intelligent identification method and system based on spatio-temporal joint perception, and belongs to the technical field of electromagnetic signal identification. Wireless channels are scanned to obtain periodically acquired signals, periodically acquired electromagnetic signals are converted into a time-frequency diagram, frequency point separation is realized by means of models such as YOLO, and the time-frequency diagram is converted into a time-frequency diagram; the method comprises the following steps: firstly, sampling a sample segment, then realizing space separation by utilizing RSSI information of the sample segment and combining a space-time joint positioning algorithm based on radio frequency fingerprints to obtain a grid space address code, and then, realizing space-time separation by constructing a data set R and utilizing a space-time joint attention mechanism to obtain an electromagnetic recognition target result; through time-space-frequency multi-dimensional separation of overlapped electromagnetic signals, accurate positioning of an electromagnetic target can be realized under a low SNR condition and a dynamic scene.

Description

Technical Field

[0001] The present invention belongs to the technical field of electromagnetic signal recognition, and specifically relates to an electromagnetic target intelligent recognition method and system based on space-time joint perception. Background Art

[0002] With the rapid development of radio technology and the popularization of equipment, the wireless electromagnetic environment is becoming increasingly complex, and the signals in the same period may be mixed with multiple types of electromagnetic signals. Accurately detecting suspicious / target signals in a certain frequency band in a complex wireless environment and timely controlling and processing them has broad application prospects in many fields.

[0003] Early electromagnetic signal recognition technologies were mainly statistical perception methods (such as energy detection, cyclostationary feature detection, matched filter detection, and covariance matrix detection methods). Their detection accuracy and false alarm rate deteriorated significantly under low signal-to-interference ratio conditions. With the development of technology, machine learning-related technologies such as decision trees and support vector machines have been applied in the field of electromagnetic signal recognition. They extract multi-dimensional features such as signal frequency domain and time domain for pattern classification and correlation analysis, significantly improving the performance of electrical signal recognition. However, since the recognition of electromagnetic signals depends on feature extraction and processing, it has a large workload. In recent years, deep learning technology has gradually become the mainstream research direction of electromagnetic signal recognition, thanks to its excellent automatic feature extraction and expression capabilities. By automatically mining the inherent laws and deep features in massive, high-dimensional samples, the performance and efficiency of this technology have been greatly improved compared to traditional technologies.

[0004] At present, deep learning electromagnetic identification methods still face many challenges: on the one hand, they are due to the common difficulties of deep learning technologies, such as the need for massive signal samples for model training, poor generalization, parameter redundancy, and lack of samples; on the other hand, they are due to the complexity and dynamics of electromagnetic signals. The electromagnetic spectrum environment is an open space where signals from various communication systems, radars, and other electronic equipment overlap. How to separate different types of electromagnetic signals mixed in unknown wireless signals is one of the current challenges. In addition, due to factors such as spatial selective fading of wireless channels, shadows, hidden terminals, and the mobility of target devices, the SNR of electromagnetic signals used for electronic detection is low, resulting in a sharp decline in the performance of existing electromagnetic identification methods.

[0005] In addition, existing electromagnetic recognition methods often process from a single signal source (such as the main signal of cognitive radio), a single feature dimension (such as energy, correlation characteristics) and a single node (difficult to eliminate the impact of multipath, etc.), without comprehensive consideration from a systematic perspective. For example, most image-based classification methods often convert electromagnetic signal data at a certain moment into a 2D image, and then use models such as CNN / YOLO for recognition, but they do not fully consider the characteristic information of electromagnetic signals in the time dimension. The time series classification method converts the electromagnetic signal data into a time series, fully mining the time-related characteristic information of the signal, but faces the problem of insufficient utilization of spatial correlation. In response to these problems, electromagnetic recognition methods based on multi-point collaboration have emerged. For example, the Chinese patent ZL202411463040.X that we have applied for discloses a collaborative electromagnetic signal intelligent recognition method and system. The method integrates the list of potential electromagnetic targets reported by each collaborative sensing node. The method comprehensively considers the status, time-frequency position consistency, graphic morphology consistency, etc. of each collaborative sensing node and performs weighted processing, thereby improving the accuracy of the recognition results. However, it still fails to solve the problem of multiple signal overlaps, and has problems of poor recognition performance and insufficient adaptability under low SNR conditions and dynamic scene conditions.

[0006] In view of this, the prior art needs to be further improved. Summary of the invention

[0007] In view of the shortcomings of the prior art, the present invention provides a method and system for intelligent recognition of electromagnetic targets based on joint space-time perception, which is committed to solving the problem of overlapping and separation of electromagnetic signals in complex space, improving the recognition performance of target electromagnetic signals under low SNR conditions and the adaptability in dynamic scenes.

[0008] In order to achieve the above technical objectives, the present invention adopts the following technical solutions: A method for intelligent recognition of electromagnetic targets based on spatiotemporal joint perception is implemented by means of an electromagnetic target intelligent recognition system, wherein the electromagnetic target intelligent recognition system comprises at least a joint processing and recognition unit capable of information exchange and a plurality of collaborative sensing nodes, and the method comprises the following steps: Step S1, the collaborative sensing node scans the wireless channel, collects electromagnetic signals at a time interval of period T, and obtains respective periodic collection signals; Step S2: Each collaborative sensing node performs frequency separation on the periodic acquisition signal obtained by each node to form electromagnetic signal information indexed by the frequency point. Step S3, the collaborative sensing node transmits the electromagnetic signal information indexed by the frequency point obtained by frequency point separation to the joint processing and identification unit. At the same time, the collaborative sensing node calculates the RSSI information of the time-frequency diagram sampling segment of each frequency point and uploads it to the joint processing and identification unit. The joint processing and identification unit first performs spatial separation on the corresponding electromagnetic signal information indexed by the frequency point according to the obtained RSSI information to obtain the electromagnetic target node position, and then performs grid space address encoding on the electromagnetic target based on the electromagnetic target node position. The joint processing and identification unit aggregates the electromagnetic signal information indexed by the frequency point and the grid space address encoding information uploaded by each collaborative sensing node, and constructs a data set R of the time-space-frequency 3D index in real time; Step S4: The joint processing recognition unit obtains the final electromagnetic target recognition result through spatiotemporal joint attention recognition based on the spatiotemporal 3D indexed data set R constructed in step S3.

[0009] Compared with the prior art, the present invention has at least the following beneficial effects: In order to solve the problem of overlapping and separating electromagnetic signals in complex space, the present invention proposes an intelligent electromagnetic target recognition method based on space-time joint perception. For the collected electromagnetic signals, they are first converted into time-frequency diagrams, and then frequency point separation is achieved with the help of models such as YOLO. Then, the RSSI information of the sampling fragments is combined with the space-time joint positioning algorithm based on radio frequency fingerprint to achieve spatial separation and obtain the grid space address code. Then, by constructing a data set R, the space-time joint attention mechanism is used to achieve space-time separation and obtain the electromagnetic recognition target result. By separating the overlapping electromagnetic signals in the "time-space-frequency" multi-dimensional way, the recognition performance under low SNR conditions and the adaptability in dynamic scenarios can be improved, so as to achieve accurate positioning of mobile electromagnetic targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings: Figure 1 It is a schematic diagram of the overall architecture of the electromagnetic target intelligent recognition system based on time-space joint perception of the present invention; Figure 2 It is a recognition flow chart of the electromagnetic target intelligent recognition method based on time-space joint perception of the present invention; Figure 3 It is a sampling timing diagram of the data acquisition process of the present invention; Figure 4 It is a schematic diagram of the process of frequency point separation of the present invention; Figure 5 is a geographic space grid layout diagram constructed by the joint processing identification unit of the present invention; Figure 6It is a schematic flow chart of the multi-channel radio frequency fingerprint joint positioning algorithm used in the present invention; Figure 7 It is a schematic structural diagram of the spatio-temporal multi-channel graph neural network model of the present invention. Detailed implementation manners

[0011] To make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in conjunction with embodiments. The illustrative embodiments of the present invention and their descriptions are only used to explain the present invention and are not intended to limit the present invention.

[0012] In view of the characteristics of wireless electromagnetic signal heterogeneity, multi-dimensional spatial correlation, diverse sources and incompleteness, etc., from a systematic perspective, the present invention proposes a deep learning technical solution based on time-space-frequency joint perception. By separating overlapping electromagnetic signals in multiple dimensions of "time-space-frequency", intelligent identification of electromagnetic targets is achieved under low SNR conditions and dynamic scenario conditions.

[0013] Specifically, as Figures 1 to 7 shown, the present invention provides an intelligent electromagnetic target identification method based on spatio-temporal joint perception, which is realized by means of an intelligent electromagnetic target identification system. The intelligent electromagnetic target identification system at least includes a joint processing and identification unit capable of information interaction and a plurality of cooperative perception nodes (see Figure 1 , the number of cooperative perception nodes is at least two, Figure 1 exemplarily shows three), and its method specifically includes the following steps: Step S1: The cooperative perception nodes scan the wireless channel, collect electromagnetic signals at time intervals of period T, and obtain their respective periodic acquisition signals; Specifically, in this step S1, the periodic acquisition signal is embodied in the form of a digital sequence. The sampling duration of each period T is Td, and Td < T. The length of the digital sequence corresponding to each sampling duration Td is p, and the formed digital sequence can be expressed as: ; In the formula, k is the number of periodic samplings. Each sampling duration Td corresponds to p digital information. For example, represents that the first sampling period T1 obtains p corresponding digital information, and the p-th digital information among them is , represents the k-th sampling period, and both p and k are positive integers; Preferably, the start time of each sampling duration Td is the start time of the corresponding sampling period T (see Figure 3 ), so as to ensure the synchronization of the sampling periods of each cooperative perception node; In addition, in step S1, the data acquisition process includes but is not limited to operations such as filtering, down-conversion and A / D conversion. The sampling timing diagram is shown in FIG. Figure 3 shown.

[0014] Step S2: Each collaborative sensing node performs frequency separation on the periodic acquisition signal obtained by each node to form electromagnetic signal information indexed by the frequency point. Specifically, for the obtained periodic acquisition signal, the corresponding cooperative sensing node first converts it into a time-frequency diagram using short-time Fourier transform. Preferably, the conversion is performed using the following formula: ; In the formula, is the DFT (Discrete Fourier Transform) of the windowed data at time m, is the nth digital information in the formed digital sequence, n∈{1,2,…,p}, is the window function, f is the frequency, j is the imaginary unit, and p is the length of the digital sequence corresponding to the sampling time Td; Next, the corresponding collaborative sensing node uses a target detection network model such as YOLO to analyze the time-frequency graph, identifies potential electromagnetic target information from the time-frequency graph, and divides the time-frequency graph into a series of time-frequency graph sampling segments indexed by frequency based on the identified potential electromagnetic target information. For each frequency point of the time-frequency graph sampling segment, a short-time inverse Fourier transform is performed one by one to form electromagnetic signal information indexed by frequency point, such as Figure 4 shown.

[0015] Step S3, the collaborative sensing node transmits the electromagnetic signal information indexed by the frequency point obtained by frequency point separation to the joint processing and identification unit. At the same time, the collaborative sensing node calculates the RSSI information of the time-frequency diagram sampling segment of each frequency point and uploads it to the joint processing and identification unit. The joint processing and identification unit serves as a data fusion center. It first performs spatial separation on the corresponding electromagnetic signal information indexed by the frequency point according to the obtained RSSI information to obtain the electromagnetic target node position, and then performs grid space address encoding on the electromagnetic target based on the electromagnetic target node position. The joint processing and identification unit aggregates the electromagnetic signal information indexed by the frequency point and the grid space address encoding information uploaded by each collaborative sensing node, and constructs a data set R of the time-space-frequency 3D index in real time; In a preferred embodiment, in step S3, in order to obtain the node position of the electromagnetic target, the joint processing and identification unit adopts a spatiotemporal joint positioning method based on radio frequency fingerprint to achieve. Specifically, the spatiotemporal joint positioning method based on radio frequency fingerprint includes the following steps: S3.1. The joint processing recognition unit takes its own position F0 as the center and constructs a radio frequency map using a nine-square grid or similar form with 2D grid coding (see Figure 5 As shown in the figure, it can also be called a geographic space grid layout diagram. Note: the number of grids and the size of each grid can be adjusted according to the actual scene); S3.2. Based on the constructed radio frequency map, the basic positioning model is obtained through offline training and the final location result is obtained: S3.21. Move the test device (transmitter target) used to obtain the RF fingerprint to each grid position of the RF map. Conduct traversal; S3.22. Collaborative sensing nodes ( m is the collaborative sensing node index, 0, 1, 2, ...; illustratively, the first collaborative sensing node is S 0 The digital sequence information (from the test equipment) received in multiple T cycles (each cycle Td) is used to calculate the corresponding RSS data and AOA data and construct a data set ,in, , i is the time period index; S3.23. The data set corresponding to each collaborative sensing node Send it to RNN / LSTM and other time-series neural networks to extract the features of multi-channel signals , this feature is a complex number, including real and imaginary parts; S3.24. Each feature The imaginary and real parts of the data are combined, and the features of multiple collaborative sensing nodes are superimposed (the combination and superposition here can be through multiplication, or simply data stacking, or other methods), to construct a multi-channel two-dimensional feature map (see for example Figure 6 ); S3.25, the multi-channel two-dimensional feature map is subjected to the SE channel attention mechanism, and the squeeze and excitation operations are performed in sequence to obtain the weight of each channel (essentially the received information of each collaborative sensing node), and then these weights are weightedly calculated with the multi-channel two-dimensional feature map (for example, multiplying each channel by channel) to obtain the weighted final feature representation; S3.26. Use a fully connected neural network to predict and output the initial recognition result of the grid location ; S3.27. Obtain the initial recognition result of the grid location by comparing and predicting And each grid position obtained by traversal (i.e., comparing the predicted value with the true value), using the LSME criterion or loss functions such as cross entropy combined with optimization algorithms such as stochastic gradient descent (SGD) to train the neural network model (the model here can be a deep learning neural network model such as CNN, RNN, Transformer, etc.) to obtain the basic positioning model (whose output is the position result); S3.28, the output result of the basic positioning model is further combined with the Kalman filter through the Adaboost mechanism to obtain the final position result output. The Adaboost mechanism and Kalman filter are used here to eliminate the uncertain effects such as data collection noise by using historical position information; S3.3. Online real-time acquisition of the node position of the electromagnetic target: S3.31. Collaborative sensing nodes Online real-time acquisition from electromagnetic targets (e.g., moving electromagnetic targets, Figure 1 The wireless signal data of the radio target 1 or 2 shown in the figure is then used to calculate the corresponding RSS data and AOA data using the digital sequence information (from the electromagnetic target) received in multiple T cycles (each cycle Td), and then splice it into the data set , get the spliced ​​dataset ; S3.32. Concatenate the datasets It is sent to the basic positioning model and processed (refer to S3.28) to obtain the monitored target position, which is the node position of the electromagnetic target.

[0016] Step S4, the joint processing recognition unit obtains the final electromagnetic target recognition result through the spatiotemporal joint attention recognition according to the spatiotemporal 3D indexed data set R constructed in step S3; Specifically, in step S4, the joint processing recognition unit performs spatiotemporal joint attention recognition by constructing a spatiotemporal multi-channel graph neural network model, wherein the spatiotemporal multi-channel graph neural network model adopts a hierarchical network structure, the underlying graph is constructed in a dynamic graph manner, and a graph attention mechanism neural network is used to extract complex spatial dependencies based on information such as position, and input it into the upper-layer time domain sequence attention model in real time to extract feature information in the time dimension, and finally output the judgment result through the fully connected layer, such as Figure 7 shown.

[0017] More specifically, the spatiotemporal multi-channel graph neural network model mainly includes a data construction layer, a spatial graph attention layer, a temporal sequence attention layer, and a fully connected layer, where each layer adopts the following settings: (1) Data construction layer The data construction layer uses the input data to construct a spatial graph for each time period in chronological order. , ,in, is the vertex of the time period graph of t, is the corresponding edge; collaborative sensing node The wireless signal data collected in period i is expressed as: , The space graph The feature data of each vertex is ; In order to reflect the mutuality of signals in space, the adjacency matrix does not use the topological connection or spatial distance between nodes, but uses the similarity based on the path vector of the signal sending and receiving nodes for representation. Specifically, let the collaborative sensing node and the target node The spatial positions are and , then the path vector of the sending and receiving nodes is The plural form of Similarly, for collaborative sensing nodes , the path vector of the sending and receiving nodes The plural form of ; The space map Network nodes on and The adjacency matrix A of (vertices) is defined as the vector With vector The similarity of , then the new adjacency matrix Characterized as the product of vector correlation and vector length difference, we get: ; In the formula, the whole before the symbol "*" on the right side is the first term, and the whole after the symbol "*" is the second term. The first term represents the vector With vector The angle correlation of With vector The relative magnitude of the magnitude; if the vector With vector In the same direction, the first term reaches the maximum value of 1; if With vector The magnitudes are the same, and the second term reaches the maximum value of 1; (2) Spatial Graph Attention Layer Compared with the existing graph attention mechanism, this paper adopts a new adjacency matrix It replaces the traditional adjacency matrix and combines it with the GCN mechanism to form a new graph attention mechanism. This mechanism not only saves the computational complexity of the existing graph attention mechanism, but also introduces the correlation between vertex features into the GCN model. Accordingly, the feature update can be expressed as: , In the formula, H is the parameter, ko is the number of updates, which means the koth update. Each calculation of parameter H must be updated. represents the degree matrix of the collaborative sensing node; σ is the nonlinear activation function; A is the adjacency matrix, which is composed of Elements; W is the weight parameter; because the perception matrix of all collaborative sensing nodes under the same fusion center is the same, can be directly omitted and the simplified expression is ; (3) Time-domain sequence attention layer The input time series at different times are correlated, and the correlation at different times is also different. Therefore, this layer uses the temporal attention mechanism to calculate the temporal dependency and extracts features using the following formula: : ; In the formula, is the query vector, is the query weight matrix, is the input vector; is the key vector, is the key weight matrix, is a value vector, is the value weight matrix; is the dimension of the vector; (4) Fully connected layer The feature information (complex number) obtained by joint spatiotemporal processing is used to obtain the judgment category through the fully connected layer and softmax function, and the judgment result is output.

[0018] The joint processing recognition unit performs spatiotemporal joint attention recognition based on the spatiotemporal-frequency 3D indexed dataset R constructed in step S3 with the help of the above spatiotemporal multi-channel graph neural network model, thereby identifying the final electromagnetic target recognition result.

[0019] It should be understood that the present invention also relates to an electromagnetic target intelligent identification system based on spatiotemporal joint perception, wherein the electromagnetic target intelligent identification system includes at least a joint processing and identification unit capable of information exchange and multiple collaborative sensing nodes, wherein the collaborative sensing nodes are mainly responsible for the collection and frequency separation of electromagnetic signals; the joint processing and identification unit is responsible for the spatial separation of electromagnetic signals and realizes spatiotemporal separation based on a spatiotemporal multi-channel graph neural network model, thereby ultimately realizing the separation and identification of complex spatial electromagnetic signals.

[0020] In order to better achieve the purpose of the present invention, the joint processing and identification unit also applies the final electromagnetic target identification result to spatial separation to optimize the accuracy of target positioning in subsequent links.

[0021] It should be noted that the above-described specific implementation methods can enable those skilled in the art to more fully understand the invention, but do not limit the invention in any way. Therefore, although the present specification and embodiments have described the invention in detail, those skilled in the art should understand that the invention can still be modified or replaced by equivalents; and all technical solutions and improvements that do not deviate from the spirit and scope of the invention are included in the protection scope of the patent for the invention. Any figure mark in the claims should not be regarded as limiting the claims involved.

Claims

1. An electromagnetic target intelligent recognition method based on spatiotemporal joint perception, the method is implemented with the aid of an electromagnetic target intelligent recognition system, the electromagnetic target intelligent recognition system at least comprising a joint processing recognition unit capable of information exchange and a plurality of collaborative sensing nodes, characterized in that: The method includes the following steps: Step S1: The cooperative perception nodes scan the wireless channel, collect electromagnetic signals at time intervals of period T, and obtain their respective periodic acquisition signals; Step S2: Each cooperative perception node separately performs frequency point separation on its obtained periodic acquisition signal to form electromagnetic signal information indexed by frequency points; Step S3: The cooperative perception nodes transmit the electromagnetic signal information indexed by frequency points obtained by frequency point separation to the joint processing and recognition unit. At the same time, the cooperative perception nodes calculate the RSSI information of the time-frequency map sampling segments of each frequency point and upload it to the joint processing and recognition unit. The joint processing and recognition unit first performs spatial separation on the corresponding electromagnetic signal information indexed by frequency points according to the obtained RSSI information to obtain the positions of electromagnetic target nodes, and then performs grid space address coding on the electromagnetic targets based on the positions of the electromagnetic target nodes. The joint processing and recognition unit aggregates the electromagnetic signal information indexed by frequency points and the grid space address coding information uploaded by each cooperative perception node to construct a data set R of a spatio-temporal-frequency 3D index in real time; Step S4: The joint processing and recognition unit obtains the final electromagnetic target recognition result through spatio-temporal joint attention recognition according to the data set R of the spatio-temporal-frequency 3D index constructed in Step S3.

2. The electromagnetic target intelligent recognition method based on time-space joint perception according to claim 1 is characterized in that: In Step S1, the periodic acquisition signal is represented in the form of a digital sequence. The sampling duration of each period T is Td, and Td < T. The length of the digital sequence corresponding to each sampling duration Td is p, and the formed digital sequence is expressed as: ； In the formula, k is the number of periodic sampling, and each sampling time Td corresponds to p digital information, among which the pth digital information is expressed as , represents the kth sampling period, where p and k are both positive integers.

3. The electromagnetic target intelligent recognition method based on time-space joint perception as claimed in claim 1 is characterized in that: In Step S2, for the obtained periodic acquisition signal, the corresponding cooperative perception node first converts it into a time-frequency map by using the short-time Fourier transform. Then, the corresponding cooperative perception node uses the target detection network model to analyze the time-frequency map, identifies potential electromagnetic target information from the time-frequency map, and based on the identified potential electromagnetic target information, divides the time-frequency map into a series of time-frequency map sampling segments indexed by frequency. For each time-frequency map sampling segment of each frequency point, the inverse short-time Fourier transform is performed one by one to form electromagnetic signal information indexed by frequency points.

4. The electromagnetic target intelligent recognition method based on time-space joint perception according to claim 1 is characterized in that: In Step S3, in order to obtain the node positions of the electromagnetic targets, the joint processing and recognition unit uses a spatio-temporal joint positioning method based on radio frequency fingerprints to achieve this. Among them, the spatio-temporal joint positioning method based on radio frequency fingerprints includes the following steps: S3.1: The joint processing and recognition unit takes its own location F0 as the center and constructs a radio frequency map with 2D grid coding; S3.2: Based on the constructed radio frequency map, obtain the basic positioning model through offline training and obtain the final position result; S3.3: Obtain the node positions of electromagnetic targets in real time online.

5. The electromagnetic target intelligent recognition method based on time-space joint perception as claimed in claim 4 is characterized in that: Step S3.2 specifically includes: S3.

21. Move the test device used to obtain the RF fingerprint to each grid position of the RF map. Conduct traversal; S3.

22. Collaborative sensing nodes Using the digital sequence information received in multiple T periods, the corresponding RSS data and AOA data are calculated and a data set is constructed. ,in, , i is the time period index; S3.

23. The data set corresponding to each collaborative sensing node Send it to the time series neural network to extract the characteristics of multi-channel signals , this feature is a complex number, including real and imaginary parts; S3.

24. Each feature The imaginary and real parts of are combined, and the features of multiple collaborative sensing nodes are superimposed to construct a multi-channel two-dimensional feature map; S3.25: Pass the multi-channel two-dimensional feature map through the SE channel attention mechanism, perform compression and excitation operations in sequence to obtain the weights of each channel, and then perform weighted calculation on these weights and the multi-channel two-dimensional feature map to obtain the weighted final feature representation; S3.

26. Use a fully connected neural network to predict and output the initial recognition result of the grid location ; S3.

27. Obtain the initial recognition result of the grid location by comparing and predicting And each grid position obtained by traversal , the basic positioning model is obtained by training the neural network model using the LSME criterion or the cross entropy loss function combined with the optimization algorithm; S3.

28. The output result of the basic positioning model is further combined with the Kalman filter through the Adaboost mechanism to obtain the final position result output.

6. The electromagnetic target intelligent recognition method based on time-space joint perception as claimed in claim 5, characterized in that: Step S3.3 specifically includes: S3.

31. Collaborative sensing nodes Collect wireless signal data from electromagnetic targets online in real time, then use the digital sequence information received in multiple T cycles to calculate the corresponding RSS data and AOA data, and splice them into the data set , get the spliced ​​dataset ; S3.

32. Concatenate the datasets The basic positioning model is sent to the basic positioning model, and its output result is further combined with the Kalman filter through the Adaboot mechanism to obtain the final position result output, which is the monitoring target position of the electromagnetic target, that is, the node position of the electromagnetic target.

7. A method for intelligent electromagnetic target recognition based on spatiotemporal joint perception according to any one of claims 1 to 6, characterized in that: In step S4, the joint processing recognition unit performs spatiotemporal joint attention recognition by constructing a spatiotemporal multi-channel graph neural network model. The spatiotemporal multi-channel graph neural network model includes a data construction layer, a spatial domain graph attention layer, a temporal sequence attention layer and a fully connected layer.

8. The electromagnetic target intelligent recognition method based on time-space joint perception as claimed in claim 7, characterized in that: In the data construction layer, the spatial graph of each time period is constructed in chronological order. In order to reflect the mutuality of signals in space, an adjacency matrix is ​​constructed. , the adjacency matrix The similarity of the signal sending and receiving node path vectors is used for characterization. in, Set up collaborative sensing nodes and the target node The spatial positions are and , then the path vector of the sending and receiving nodes is The plural form of ; For collaborative sensing nodes , the path vector of the sending and receiving nodes The plural form of ; Then, the adjacency matrix Characterized by: ； Among them, on the right side of the equal sign "=", the whole before the symbol "*" is the first term, and the whole after the symbol "*" is the second term. The first term represents the vector With vector The angle correlation of With vector The relative size of the amplitude.

9. An electromagnetic target intelligent recognition system based on time-space joint perception, which is used to implement an electromagnetic target intelligent recognition method based on time-space joint perception as described in any one of claims 7-8, characterized in that: The collaborative sensing node is responsible for the collection and frequency separation of electromagnetic signals; the joint processing and identification unit is responsible for the spatial separation of electromagnetic signals and realizes spatiotemporal separation based on the spatiotemporal multi-channel graph neural network model, thereby ultimately realizing the separation and identification of complex spatial electromagnetic signals.

10. The electromagnetic target intelligent recognition system based on time-space joint perception as claimed in claim 9, characterized in that: The joint processing and identification unit also applies the final electromagnetic target recognition results to spatial separation to optimize the accuracy of target positioning in subsequent links.

Images