Non-cooperative target side lobe sensing method and system, medium and equipment

Abstract

The invention discloses a non-cooperative target sidelobe sensing method and system, a medium and equipment, and belongs to the technical field of spectrum sensing. The method comprises the following steps: acquiring a to-be-sensed sidelobe signal emitted by a non-cooperative target; establishing a cooperative sensing problem of to-be-sensed sidelobe signals under different signal-to-noise ratios; inputting the to-be-sensed sidelobe signal into a pre-constructed sensing network model, and outputting a sidelobe sensing result of the non-cooperative target; wherein the processing step of the sensing network model comprises the following steps: according to the cooperative sensing problem, judging a sensing result of a to-be-sensed sidelobe signal at a sensing node; and deciding a sidelobe sensing result of the non-cooperative target according to a sensing result of the to-be-sensed sidelobe signal at the sensing node. According to the method, the cooperative sensing problem of the to-be-sensed sidelobe signals under different signal-to-noise ratios is established, and the sensing result is judged and decided through the cooperative sensing problem, so that the accuracy of detecting the non-cooperative sidelobe signals is greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of spectrum sensing technology, and in particular to a sidelobe sensing method, system, medium and device for a non-cooperative target. Background Art

[0002] With the rapid development of communication technology and the rapid growth of electronic communication equipment, people's demand for spectrum resources is also increasing. On the one hand, scholars develop spectrum resources from the frequency domain by adopting higher frequency bands for communication, such as using mmWave communication technology, which has ultra-wide (30-300 GHz) spectrum resources. On the other hand, scholars use beamforming technology to design high-gain directional antennas, so that the transmitter can transmit signals to the receiver from different angles, which meets the growing demand for speed and throughput and provides sufficient bandwidth for the growing spectrum demand.

[0003] When a non-cooperative target communicates with a remote receiving end where there is no direct link, it usually uses an air relay to deliver the information. In order to reduce the probability of being discovered, it is assumed that the non-cooperative target transmits information via an air relay. If we want to perceive such non-cooperative targets, the best way is to capture the signal between the target and the air relay uplink. However, due to the difficulty in finding the uplink in the absence of prior information, the target information cannot be perceived. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and to provide a sidelobe sensing method, system, medium and device for a non-cooperative target, which utilizes the sidelobe signal and leakage signal of the target to perceive the target information, and through the perception network model, greatly improves the accuracy of detecting non-cooperative sidelobe signals.

[0005] To achieve the above object, the present invention is implemented by adopting the following technical solutions:

[0006] On the one hand, the present invention provides a sidelobe sensing method for a non-cooperative target, comprising:

[0007] Acquire the side lobe signal to be sensed emitted by the non-cooperative target;

[0008] Establish cooperative sensing problem of sidelobe signals to be sensed under different signal-to-noise ratios;

[0009] Inputting the sidelobe signal to be sensed into a pre-built perception network model, and outputting the sidelobe perception result of the non-cooperative target;

[0010] The processing steps of the perception network model include:

[0011] According to the collaborative sensing problem, determining the sensing result of the sidelobe signal to be sensed at the sensing node;

[0012] According to the perception result of the sidelobe signal to be sensed at the sensing node, the sidelobe perception result of the non-cooperative target is determined.

[0013] Optionally, the step of acquiring the to-be-sensed sidelobe signal emitted by the non-cooperative target includes:

[0014] According to the angle between the main lobe beam transmitted by the non-cooperative target and the antenna, the transmitting main lobe beamforming vector, the receiving main lobe beamforming vector and the main lobe channel gain vector of the non-cooperative target are calculated;

[0015] Calculate the main lobe signal transmitted by the non-cooperative target according to the transmit main lobe beamforming vector, the receive main lobe beamforming vector and the main lobe channel gain vector;

[0016] According to the angle between the sidelobe beam emitted by the non-cooperative target and the antenna, the transmit sidelobe beamforming vector, the receive sidelobe beamforming vector and the sidelobe channel gain vector of the non-cooperative target are calculated;

[0017] The sidelobe signal to be sensed transmitted by the non-cooperative target is calculated according to the transmit sidelobe beamforming vector, the receive sidelobe beamforming vector, the sidelobe channel gain vector and the mainlobe signal.

[0018] Optionally, the main lobe signal transmitted by the non-cooperative target is expressed as:

[0019] ;

[0020] ;

[0021] ;

[0022] ;

[0023] ;

[0024] in, Represents the main lobe signal transmitted by a non-cooperative target; represents the main lobe steering vector; represents additive noise that follows a zero-mean additive Gaussian white noise distribution; represents the output of the antenna; represents the receive main lobe beamforming vector of the non-cooperative target; represents the transmit mainlobe beamforming vector of the non-cooperative target; represents the main lobe channel gain vector; represents the imaginary part; Indicates the spacing between transmitting antennas; Represents the angle between the main lobe beam emitted by the non-cooperative target and the antenna; represents the number of receiving antennas of non-cooperative targets; represents the number of transmit antennas of non-cooperative targets; It represents the reflection coefficient of the main lobe beam pointing to the ground; Indicates the carrier wavelength; Represents the distance from the first antenna to the sensing node.

[0025] Optionally, the sidelobe signal to be sensed emitted by the non-cooperative target is expressed as:

[0026] ;

[0027] ;

[0028] ;

[0029] ;

[0030] ;

[0031] ;

[0032] in, represents the sidelobe signal to be sensed emitted by the non-cooperative target; represents the transmit power of the non-cooperative target; represents the sidelobe steering vector; represents a hyperparameter; represents the receive sidelobe beamforming vector of the non-cooperative target; represents the transmit sidelobe beamforming vector of the non-cooperative target; represents the sidelobe channel gain vector; represents the angle between the sidelobe beam emitted by the non-cooperative target and the antenna; , They represent the azimuth and elevation angles of the sidelobe signal respectively.

[0033] Optionally, the collaborative sensing problem of the sidelobe signals to be sensed under different signal-to-noise ratios is expressed as:

[0034] ;

[0035] in, It means that it follows zero-mean additive Gaussian white noise; represents the transmit power of the non-cooperative target; represents the sidelobe steering vector; represents the main lobe steering vector; represents a hyperparameter; represents the output of the antenna; Indicates that the sidelobe signal to be sensed does not exist; Indicates that the sidelobe signal to be sensed exists.

[0036] Optionally, the perception network model includes a perception node network and a fusion center network, and the output layer of the perception node network is connected to the input layer of the fusion center network;

[0037] The perception node network and the fusion center network have the same structure, both including an input layer, a convolution layer, a residual module, a normalization layer, a fully connected layer, an activation function layer and an output layer connected in sequence.

[0038] Optionally, the processing steps of the perception network model include:

[0039] Input the sidelobe signal to be sensed into the sensing node network, perform feature extraction and residual processing in sequence, and obtain a feature map of the sidelobe signal to be sensed; perform feature mapping on the feature map of the sidelobe signal to be sensed, and determine the perception result of the feature map of the sidelobe signal to be sensed at the sensing node according to the collaborative sensing problem;

[0040] The perception results of the perception nodes are input into the fusion center network, and secondary feature extraction, secondary residual processing, and secondary feature mapping are performed in sequence to determine the sidelobe perception results of the non-cooperative target.

[0041] In a second aspect, the present invention provides a sidelobe perception system for a non-cooperative target, comprising:

[0042] A sidelobe signal acquisition module, used to acquire the sidelobe signal to be sensed emitted by the non-cooperative target;

[0043] A perception problem establishment module, used to establish collaborative perception problems of sidelobe signals to be sensed under different signal-to-noise ratios;

[0044] A perception result calculation module, used for inputting the sidelobe signal to be perceived into a pre-built perception network model, and outputting the sidelobe perception result of the non-cooperative target;

[0045] The processing steps of the perception network model include:

[0046] According to the collaborative sensing problem, determining the sensing result of the sidelobe signal to be sensed at the sensing node;

[0047] According to the perception result of the sidelobe signal to be sensed at the sensing node, the sidelobe perception result of the non-cooperative target is determined.

[0048] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the method described in the first aspect when executed by a processor.

[0049] In a fourth aspect, the present invention provides a computer program device, including a processor and a storage medium;

[0050] The storage medium is used to store instructions;

[0051] The processor is configured to operate according to the instructions to execute the method according to the first aspect.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] The present invention establishes a collaborative perception problem of sidelobe signals to be perceived under different signal-to-noise ratios, judges and decides the perception result through the collaborative perception problem, and uses the sidelobe signals and leakage signals of the target to perceive the target information. Through the perception network model, the accuracy of detecting non-cooperative sidelobe signals is greatly improved; and it can be expanded to large-scale scenarios, effectively improving the detection performance of the perception network for sidelobe signals compared with traditional methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 FIG. 1 is a flow chart of a sidelobe sensing method for a non-cooperative target in an embodiment of the present invention; Figure 2 The figure shows a schematic diagram of the structure of a perception scene in one embodiment of the present invention; Figure 3 Shown is a schematic diagram of the workflow of the perception network model of the present invention in one embodiment; Figure 4 It is a schematic diagram of the structure of a sensing node network or a fusion center network in one embodiment of the present invention; Figure 5 FIG. 1 is a schematic diagram showing a comparison of detection performances of the perception network model of the present invention and the existing perception network model in one embodiment; Figure 6 FIG. 1 is a schematic diagram showing a comparison of detection performances of the perception network model of the present invention and the existing perception network model in another embodiment; Figure 7 The figure is a schematic diagram of the comparison of the root mean square error performance in one embodiment of the present invention; Figure 8 The figure is a schematic diagram of the comparison of the root mean square error performance in the second embodiment of the present invention; Fig. 9 FIG. 1 is a schematic diagram showing a comparison of root mean square error performance in a third embodiment of the present invention; Fig.10 FIG. 1 is a schematic diagram showing a comparison of root mean square error performance in a fourth embodiment of the present invention; Fig.11 FIG. 1 is a schematic diagram showing a comparison of root mean square error performance in a fifth embodiment of the present invention; Fig.12 FIG. 1 is a schematic diagram showing a comparison of root mean square error performance in a sixth embodiment of the present invention; Fig.13 FIG. 1 is a schematic diagram showing a comparison of root mean square error performance in a seventh embodiment of the present invention; Fig.14 FIG. 1 is a schematic diagram showing a comparison of root mean square error performance in an eighth embodiment of the present invention; Fig.15 FIG. 1 is a schematic diagram of beam estimation in an embodiment of the present invention; Fig.16 FIG. 2 is a schematic diagram of beam estimation in a second embodiment of the present invention; Fig.17 FIG. 2 is a schematic diagram of beam estimation in a third embodiment of the present invention; Fig.18 FIG. 4 is a schematic diagram of beam estimation in the fourth embodiment of the present invention. DETAILED DESCRIPTION

[0055] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. The embodiments of the present invention and the technical features in the embodiments may be combined with each other unless there is a conflict.

[0056] The term "and / or" is only a description of the association relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " generally indicates that the related objects are in an "or" relationship.

[0057] Example 1

[0058] like Figure 1 As shown, this embodiment introduces a sidelobe sensing method for a non-cooperative target. When a non-cooperative target communicates with a remote receiving end where there is no direct link, it generally uses an air relay to deliver information. In order to reduce the probability of being discovered, it is assumed that the non-cooperative target transmits information via an air relay. If we want to sense such a non-cooperative target, the most ideal method is to capture the signal of the target and the air relay uplink.

[0059] However, since it is difficult to find the uplink without prior information, we deploy a perception network model on the ground and use the target's sidelobe signal and leakage signal to perceive the target information. It is assumed that the target generates a beam signal through the array antenna and sends it to the air relay, which then transmits the signal to the user. Since the beam signal generated by the target is emitted in the air, that is, the main lobe signal is emitted into the air, the signal is not easy to be found. Therefore, our perception network model perceives the target through the ground sidelobe signal generated by the target. The perception scene is as follows: Figure 2 As shown, it includes a sidelobe signal and a power leakage signal generated when a non-cooperative target sends a signal to an upper relay node, and a perception network model deployed by us including multiple perception node networks and a fusion center network. This embodiment uses the perception network model to perceive the sidelobe signal to be perceived emitted by the non-cooperative target; the method includes the following steps:

[0060] Step 1: Obtain the sidelobe signal to be sensed emitted by the non-cooperative target, specifically:

[0061] Assuming that the antenna array of the target transmitter is a linear array, the angle between the main lobe beam emitted by the non-cooperative target and the antenna is express, ; Assuming that the path attenuation of all antennas is the same, the main lobe signal transmitted by the non-cooperative target It can be expressed as:

[0062] ;

[0063] in, represents the transmit power of the non-cooperative target; represents the imaginary part; Indicates the distance the signal is transmitted; represents the carrier wavelength, ,in, is the carrier frequency, is the speed of light; the spacing between the transmitting antennas is , the distance from the first antenna, which is the antenna closest to the receiver, to the sensing node is , the distance from the second antenna to the sensing node is , the distance from the nth antenna to the sensing node is .

[0064] According to the angle between the main lobe beam transmitted by the non-cooperative target and the antenna, the transmit main lobe beamforming vector, the receive main lobe beamforming vector and the main lobe channel gain vector of the non-cooperative target are calculated. , the receiving main lobe beamforming vector of the non-cooperative target and the main lobe channel gain vector It is expressed as:

[0065] ;

[0066] ;

[0067] ;

[0068] in, represents the imaginary part; represents the number of receiving antennas of non-cooperative targets; represents the number of transmit antennas of non-cooperative targets; , Represents the reflection coefficient of the main lobe beam pointing to the ground.

[0069] According to the transmit main lobe beamforming vector , the receiving main lobe beamforming vector of the non-cooperative target and the main lobe channel gain vector Calculate the mainlobe signal transmitted by a non-cooperative target , the main lobe signal emitted by the non-cooperative target Rewritten as:

[0070] ;

[0071] ;

[0072] in, represents the main lobe steering vector; represents additive noise following a zero-mean additive white Gaussian noise (AWGN) distribution with a variance of ; represents the output of the antenna;

[0073] According to the angle between the sidelobe beam emitted by the non-cooperative target and the antenna, the transmit sidelobe beamforming vector, the receive sidelobe beamforming vector and the sidelobe channel gain vector of the non-cooperative target are calculated, wherein the transmit sidelobe beamforming vector of the non-cooperative target is , the receive sidelobe beamforming vector of the non-cooperative target and the sidelobe channel gain vector It is expressed as:

[0074] ;

[0075] ;

[0076] ;

[0077] ;

[0078] in, represents the angle between the sidelobe beam emitted by the non-cooperative target and the antenna; , They represent the azimuth and elevation angles of the sidelobe signal respectively.

[0079] According to the transmit sidelobe beamforming vector of the non-cooperative target , the receive sidelobe beamforming vector of the non-cooperative target and the sidelobe channel gain vector , calculate the sidelobe signal to be sensed emitted by the non-cooperative target , expressed as:

[0080] ;

[0081] ;

[0082] in, represents a hyperparameter.

[0083] Step 2: Establish cooperative sensing of sidelobe signals under different signal-to-noise ratios , expressed as:

[0084] ;

[0085] Among them, the hyperparameters used to determine whether the beam signal exists are express, Indicates that the sidelobe signal to be sensed does not exist; Indicates that the sidelobe signal to be sensed exists.

[0086] Step 3: inputting the sidelobe signal to be sensed into a pre-built perception network model, and outputting the sidelobe perception result of the non-cooperative target;

[0087] Since the sidelobe signal is very weak, the performance of the general energy detection method is very poor. We perceive the target sidelobe signal by deploying a perception network model composed of multiple perception node networks and a fusion center network; the perception network model includes a perception node network and a fusion center network, and the output layer of the perception node network is connected to the input layer of the fusion center network; the perception node network and the fusion center network have the same structure, both including an input layer, a convolution layer, a residual module, a normalization layer, a fully connected layer, an activation function layer and an output layer connected in sequence, and the structure of the perception node network or the fusion center network is as follows Figure 4 shown.

[0088] The perception network model can be a BDnet model, that is, the perception node network can be a BDSNet model, and the fusion center network can be a BDFNet model. The design of the BDNet model structure is inspired by the multi-node fusion perception framework. Different from the single-point perception network model structure designed for a single node in the past, the BDNet model not only performs offline training of single-point perception for each node in the perception node network, but also performs offline training for the fusion center network to obtain the perception results uploaded by each perception node.

[0089] Specifically, the BDSNet model is used to independently perceive whether a single node in the model has a sidelobe signal, and the BDFNet model makes the final decision by collecting the perception results uploaded by all perception nodes.

[0090] The processing steps of the perception network model are as follows: Figure 3 As shown, specifically including:

[0091] The side lobe signal to be sensed is input into the BDSNet model through the input layer of the BDSNet model, and passes through the convolution layer and the residual module in sequence to perform feature extraction and residual processing to obtain a feature map of the side lobe signal to be sensed; in each convolution layer, the convolution kernel performs convolution on the side lobe signal to be sensed, i.e., feature extraction, and then a nonlinear activation function is applied to obtain the side lobe feature to be sensed, and the side lobe feature to be sensed is input into the difference block module to promote better feature learning and regularization, thereby helping to alleviate the overfitting problem and obtain a feature map of the side lobe signal to be sensed;

[0092] The feature graph of the sidelobe signal to be sensed is input into the fully connected layer for feature mapping. According to the collaborative sensing problem, the perception result of the feature graph of the sidelobe signal to be sensed at the sensing node is determined. Then, the i-th sensing node determines whether the sidelobe signal to be sensed exists. Indicates that Output through the output layer of the BDSNet model.

[0093] There are normalization layers and activation function layers before and after the fully connected layer. The activation function layer uses a modified linear rectifier unit layer. The normalization layer reduces the change of input distribution by standardizing the input data. It can accelerate the convergence process of the network and increase the training speed. We obtain standardized data by calculating the mean and variance.

[0094] The perception result of the perception node is The BDFNet model is input through its input layer, and passes through the convolution layer, residual module, normalization layer, fully connected layer, and activation function layer in sequence. Secondary feature extraction, secondary residual processing, and secondary feature mapping are performed in sequence to determine the sidelobe perception results of non-cooperative targets and output them through the output layer of the BDFNet model.

[0095] Each sensor node network uses the BDSNet model for learning: Each sensor node first collects a large amount of observation data of the sidelobe signal to be sensed, labels it, and pre-trains the BDSNet model. After that, we input the real data into the trained BDSNet model and finally obtain the judgment result of a single sensor node.

[0096] The fusion center network uses the BDFNet model for learning: all observation nodes in the perception node network feed back the perception results to the fusion center network, and the fusion center network obtains the perception results uploaded by all perception nodes such as "0011111000", and the fusion center network uses a large amount of perception result data uploaded by all labeled perception nodes for training. After that, the judgment result of each perception node is input into the trained fusion center network to obtain the final decision, that is, the sidelobe perception result of the non-cooperative target.

[0097] Example 2

[0098] This embodiment introduces a specific experimental example of a sidelobe sensing method for a non-cooperative target, including the following steps:

[0099] like Figure 2 As shown in the figure, the deployed perception network model includes multiple perception node networks and a fusion center network. It is assumed that the non-cooperative target generates a beam signal through the array antenna and sends it to the air relay, which then transmits the signal to the user. Since the beam signal generated by the target is emitted in the air, that is, the main lobe signal is emitted into the air, the signal is not easy to be detected. Therefore, the perception network model perceives the target through the ground sidelobe signal generated by the target.

[0100] Since the sidelobe signal is very weak, the performance of the general energy detection method is very poor. We use the BDnet model to perceive the target sidelobe signal in the perception network model.

[0101] The BDNet model is a fusion perception signal detection solution based on neural networks. The design of its network structure is inspired by the multi-node fusion perception framework. Different from the single-point perception network structure designed for a single perception node in the past, the BDNet model not only performs offline training for single-point perception for each perception node in the perception node network, but also performs offline training for the fusion center network to obtain the perception results uploaded by each perception node. Figure 3 As shown in the figure, the BDNet model consists of the BDSNet model and the BDFNet model.

[0102] like Figure 4As shown in the figure, both the BDSNet model and the BDFNet model consist of a convolutional layer, two residual layers, and a fully connected layer. The BDSNet model is used to independently sense whether a single sensing node in the network has a sidelobe signal. The BDFNet model collects the perception results uploaded by all sensing nodes and makes the final decision.

[0103] like Figure 5 As shown in Figure 1, in the single-point detection stage, we first compare the perceptual performance of each perceptual node of the sub-network BDSNet model of the BDNet model with other classic networks. The classic 7-layer convolutional neural network, the Visual Geometry Group (VGG) network has 16 layers and a 2-layer long short-term memory (LSTM) network. Figure 5 As shown in the figure, when the target distance is within 30m, the performance of the BDNet model is significantly better than the other three networks. When the target distance is 30~40m, the detection rate of VGG is better than that of the BDNet model.

[0104] like Figure 6 As shown in the figure, when the target distance is 30~40m, the BDNet model has a better false alarm rate than VGG. In addition, when the target distance is 30m, the detection probability of VGG is 66.1%, and the detection probability of the BDNet model is 57.8%. When the target distance exceeds 30m, the detection rate of VGG is less than 50%. Therefore, the performance of the BDSNet model is significantly better than other traditional networks.

[0105] Figure 7-14 The standardized confusion matrix of the BDSNet model is shown when the sensing node is 10-80m away from the target sidelobe. In the standardized confusion matrix, we can easily see the two key performance indicators of spectrum sensing, detection probability and false alarm probability. Figure 7 As shown, the detection probability is shown in the lower left corner of the confusion matrix, while the false alarm probability is shown in the upper right corner of the confusion matrix.

[0106] like Figure 15-18 As shown in the figure, we show the perception performance after using the BDFNet model in a perception network model using 10 perception nodes, and show the final judgment results given by the BDFNet model when the farthest perception node is at different distances from the non-cooperative target. Obviously, after using the BDFNet model, the perception performance is further improved.

[0107] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0108] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0109] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0110] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0111] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

Claims

1. A sidelobe sensing method for a non-cooperative target, characterized in that: include: Acquire the sidelobe signal to be sensed emitted by the non-cooperative target; Establish cooperative sensing problem of sidelobe signals to be sensed under different signal-to-noise ratios; Inputting the sidelobe signal to be sensed into a pre-built perception network model, and outputting the sidelobe perception result of the non-cooperative target; The processing steps of the perception network model include: According to the collaborative sensing problem, determining the sensing result of the sidelobe signal to be sensed at the sensing node; According to the perception result of the sidelobe signal to be sensed at the sensing node, the sidelobe perception result of the non-cooperative target is determined.

2. The sidelobe sensing method for a non-cooperative target according to claim 1, characterized in that: The step of acquiring the sidelobe signal to be sensed emitted by the non-cooperative target comprises: According to the angle between the main lobe beam transmitted by the non-cooperative target and the antenna, the transmitting main lobe beamforming vector, the receiving main lobe beamforming vector and the main lobe channel gain vector of the non-cooperative target are calculated; Calculate the main lobe signal transmitted by the non-cooperative target according to the transmit main lobe beamforming vector, the receive main lobe beamforming vector and the main lobe channel gain vector; According to the angle between the sidelobe beam emitted by the non-cooperative target and the antenna, the transmit sidelobe beamforming vector, the receive sidelobe beamforming vector and the sidelobe channel gain vector of the non-cooperative target are calculated; The sidelobe signal to be sensed transmitted by the non-cooperative target is calculated according to the transmit sidelobe beamforming vector, the receive sidelobe beamforming vector, the sidelobe channel gain vector and the mainlobe signal.

3. The sidelobe sensing method for a non-cooperative target according to claim 2, characterized in that: The main lobe signal emitted by the non-cooperative target is expressed as: ； ； ； ； ； in, Represents the main lobe signal transmitted by a non-cooperative target; represents the main lobe steering vector; represents additive noise that follows a zero-mean additive Gaussian white noise distribution; represents the output of the antenna; represents the receive main lobe beamforming vector of the non-cooperative target; represents the transmit mainlobe beamforming vector of the non-cooperative target; represents the main lobe channel gain vector; represents the imaginary part; Indicates the spacing between transmitting antennas; represents the angle between the main lobe beam emitted by the non-cooperative target and the antenna; represents the number of receiving antennas of non-cooperative targets; represents the number of transmit antennas of non-cooperative targets; It represents the reflection coefficient of the main lobe beam pointing to the ground; Indicates the carrier wavelength; Represents the distance from the first antenna to the sensing node.

4. The sidelobe sensing method for a non-cooperative target according to claim 3, characterized in that: The sidelobe signal to be sensed emitted by the non-cooperative target is expressed as: ； ； ； ； ； ； in, represents the sidelobe signal to be sensed emitted by the non-cooperative target; represents the transmit power of the non-cooperative target; represents the sidelobe steering vector; represents a hyperparameter; represents the receive sidelobe beamforming vector of the non-cooperative target; represents the transmit sidelobe beamforming vector of the non-cooperative target; represents the sidelobe channel gain vector; represents the angle between the sidelobe beam emitted by the non-cooperative target and the antenna; , They represent the azimuth and elevation angles of the sidelobe signal respectively.

5. The sidelobe sensing method for a non-cooperative target according to claim 1, characterized in that: The cooperative sensing problem of the sidelobe signals to be sensed under different signal-to-noise ratios is expressed as: ； in, It means that it follows zero-mean additive Gaussian white noise; represents the transmit power of the non-cooperative target; represents the sidelobe steering vector; represents the main lobe steering vector; represents a hyperparameter; represents the output of the antenna; Indicates that the sidelobe signal to be sensed does not exist; Indicates that the sidelobe signal to be sensed exists.

6. The sidelobe sensing method of a non-cooperative target according to claim 1, characterized in that: The perception network model includes a perception node network and a fusion center network, and the output layer of the perception node network is connected to the input layer of the fusion center network; The perception node network and the fusion center network have the same structure, both including an input layer, a convolution layer, a residual module, a normalization layer, a fully connected layer, an activation function layer and an output layer connected in sequence.

7. The sidelobe sensing method for a non-cooperative target according to claim 6, characterized in that: The processing steps of the perception network model include: Input the sidelobe signal to be sensed into the sensing node network, perform feature extraction and residual processing in sequence, and obtain a feature map of the sidelobe signal to be sensed; perform feature mapping on the feature map of the sidelobe signal to be sensed, and determine the perception result of the feature map of the sidelobe signal to be sensed at the sensing node according to the collaborative sensing problem; The perception results of the perception nodes are input into the fusion center network, and secondary feature extraction, secondary residual processing, and secondary feature mapping are performed in sequence to determine the sidelobe perception results of the non-cooperative target.

8. A sidelobe sensing system for a non-cooperative target, characterized in that: include: A sidelobe signal acquisition module, used to acquire the sidelobe signal to be sensed emitted by the non-cooperative target; A perception problem establishment module, used to establish collaborative perception problems of sidelobe signals to be sensed under different signal-to-noise ratios; A perception result calculation module, used for inputting the sidelobe signal to be perceived into a pre-built perception network model, and outputting the sidelobe perception result of the non-cooperative target; The processing steps of the perception network model include: According to the collaborative sensing problem, determining the sensing result of the sidelobe signal to be sensed at the sensing node; According to the perception result of the sidelobe signal to be sensed at the sensing node, the sidelobe perception result of the non-cooperative target is determined.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

10. A computer program device, characterized in that: including processor and storage medium; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the method according to any one of claims 1 to 7.

Images