A radar seeker anti-jamming method based on integrated detection and tracking

Abstract

The application provides a radar seeker anti-interference method based on integrated detection and target detection, which integrates detection, interference identification, countermeasures and target detection processing functions to improve the target detection performance of the radar seeker. The method comprises the following steps: obtaining a to-be-processed echo signal, performing feature extraction and fusion on the to-be-processed echo signal to obtain the fusion features of the to-be-processed echo signal; inputting the fusion features of the to-be-processed echo signal into an interference identification model to obtain the category of the interference signal in the to-be-processed echo signal; generating an interference situation map based on the category of the interference signal in the to-be-processed echo signal and other known information, and then selecting the optimal anti-interference processing mode for the to-be-processed echo signal from a dynamic strategy library by using a Q-learning game decision method based on the interference situation map; processing the to-be-processed echo signal by using the optimal anti-interference processing mode to obtain an echo signal after anti-interference processing, and then performing target detection to identify and locate the position of the target.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of radar, in particular to a radar seeker anti-interference method, system and device based on detection-interference-probe integration and a storage medium. BACKGROUND

[0002] Radar seekers play an important role in modern military and civilian fields. However, in a complex electromagnetic environment, radar seekers are vulnerable to various interferences (such as electronic interference, noise interference, deception interference, etc.), which can lead to a decrease in target detection accuracy or even failure. Traditional anti-interference techniques usually use a single interference detection or suppression method, which is difficult to cope with complex and variable interference environments.

[0003] In addition, in the prior art, detection, interference identification and anti-interference processing are usually independent modules, each of which undertakes different tasks. However, since the processing of each module is independent, the system response speed is slow and the real-time performance is poor. In the face of a dynamically changing interference environment, the detection and identification process may lag behind the actual interference changes, resulting in the anti-interference processing being unable to start in time, which in turn affects the overall performance of the system.

[0004] Therefore, there is an urgent need for an efficient anti-interference technology that integrates detection, interference identification and countermeasures with target detection processing. SUMMARY

[0005] The embodiments of the present application provide a radar seeker anti-interference method based on detection-interference-probe integration, which integrates detection, interference identification and countermeasures with target detection processing functions to achieve efficient and real-time anti-interference capability and improve the target detection performance of radar seekers in a complex electromagnetic environment.

[0006] To achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0007] In a first aspect, the present application provides a radar seeker anti-interference method based on detection-interference-probe integration, which comprises:

[0008] Obtaining a to-be-processed echo signal and performing feature extraction and fusion on the to-be-processed echo signal to obtain a fusion feature of the to-be-processed echo signal;

[0009] Inputting the fusion feature of the to-be-processed echo signal into an interference identification model to obtain the category of interference signals in the to-be-processed echo signal;

[0010] Based on the category of interference signals in the to-be-processed echo signal, the known position information of the radar seeker when the to-be-processed echo signal is collected and the known direction information of the interference source when the to-be-processed echo signal is collected, an interference situation map is generated;

[0011] Based on the interference situation map, a state space of the radar seeker at any time is obtained;

[0012] Based on the state space of the radar seeker at any moment and a plurality of anti-interference processing modes in the dynamic strategy library, a reward function is set to determine reward function values corresponding to the plurality of anti-interference processing modes executed by the radar seeker, wherein the dynamic strategy library is preset;

[0013] Based on the reward function values corresponding to the plurality of anti-interference processing modes executed by the radar seeker, the state space of the radar seeker at any moment and the plurality of anti-interference processing modes in the dynamic strategy library, a Q-learning algorithm is used to calculate expected returns corresponding to the plurality of anti-interference processing modes executed by the radar seeker, and then an anti-interference processing mode corresponding to a maximum expected return is selected as an optimal anti-interference processing mode for the to-be-processed echo signal;

[0014] The to-be-processed echo signal is subjected to anti-interference processing by using the optimal anti-interference processing mode, to obtain an anti-interference-processed echo signal, and then target detection is performed on the anti-interference-processed echo signal to identify and locate a position of the target.

[0015] In a second aspect, the present application provides a radar seeker anti-interference system based on detection and jamming integration, which comprises:

[0016] An acquisition module is configured to acquire a to-be-processed echo signal;

[0017] A feature extraction and fusion module is configured to perform feature extraction and fusion on the to-be-processed echo signal to obtain fusion features of the to-be-processed echo signal;

[0018] A model processing module is configured to input the fusion features of the to-be-processed echo signal into an interference identification model to obtain a category of interference signals in the to-be-processed echo signal;

[0019] The interference situation assessment and intelligent decision module is configured to generate an interference situation diagram based on the category of the interference signal in the to-be-processed echo signal, known position information of the radar seeker at the time of collection of the to-be-processed echo signal, and known direction information of the interference source at the time of collection of the to-be-processed echo signal, and obtain a state space of the radar seeker at any moment based on the interference situation diagram; the interference situation assessment and intelligent decision module is further configured to determine reward function values corresponding to the radar seeker executing a plurality of anti-interference processing modes respectively by setting a reward function based on the state space of the radar seeker at any moment and the plurality of anti-interference processing modes in the dynamic strategy library; the interference situation assessment and intelligent decision module is further configured to calculate expected returns corresponding to the radar seeker executing the plurality of anti-interference processing modes respectively by a Q-learning algorithm based on the reward function values corresponding to the radar seeker executing the plurality of anti-interference processing modes respectively, the state space of the radar seeker at any moment, and the plurality of anti-interference processing modes in the dynamic strategy library, and select an anti-interference processing mode corresponding to a maximum expected return as an optimal anti-interference processing mode for the to-be-processed echo signal, wherein the dynamic strategy library is preset.

[0020] The detection and tracking module is configured to perform anti-interference processing on the to-be-processed echo signal by using the optimal anti-interference processing mode to obtain an anti-interference processed echo signal, and perform target detection on the anti-interference processed echo signal to identify and locate the position of the target.

[0021] In a third aspect, an anti-interference device for a radar seeker based on detection and interference exploration integration is provided. The anti-interference device includes modules for executing the method of the first aspect.

[0022] In a possible design, the anti-interference device for the radar seeker based on detection and interference exploration integration of the third aspect can further include a transceiver. The transceiver can be a transceiver circuit or an interface circuit. The transceiver can be configured to enable the anti-interference device for the radar seeker based on detection and interference exploration integration of the third aspect to communicate with other devices.

[0023] In a possible design, the anti-interference device for the radar seeker based on detection and interference exploration integration of the third aspect can further include a memory. The memory can be integrated with the processor or can be separately arranged. The memory can be configured to store instructions related to the method of the first aspect.

[0024] In a fourth aspect, an anti-interference device for a radar seeker based on detection and interference exploration integration is provided. The anti-interference device includes a processor coupled with a memory. The processor is configured to execute instructions stored in the memory to enable the anti-interference device for the radar seeker based on detection and interference exploration integration to execute the method of the first aspect.

[0025] In one possible design, the anti-jamming device for the radar seeker based on integrated detection and interception in the fourth aspect may further include a transceiver. This transceiver can be a transceiver circuit or an interface circuit. The transceiver can be used for communication between the anti-jamming device for the radar seeker based on integrated detection and interception in the fourth aspect and other devices.

[0026] Fifthly, a radar seeker anti-jamming device based on integrated detection and interception is provided, comprising: a processor and a memory; the memory is used to store instructions, and when the processor executes the instructions, the radar seeker anti-jamming device based on integrated detection and interception performs the method of the first aspect.

[0027] In one possible design, the anti-jamming device for the radar seeker based on integrated detection and interception in the fifth aspect may further include a transceiver. This transceiver can be a transceiver circuit or an interface circuit. The transceiver can be used for communication between the anti-jamming device for the radar seeker based on integrated detection and interception in the fifth aspect and other devices.

[0028] In a sixth aspect, a computer-readable storage medium is provided, the computer-readable storage medium including storage of a computer program or instructions, which, when executed, cause the anti-jamming method of the radar seeker based on the first aspect of detection and interception to be performed.

[0029] In this embodiment, a closed-loop system of "detection-interference identification-detection" is formed by designing multi-domain feature joint extraction, spatiotemporal-frequency three-domain joint filtering, deep learning classification model, interference situation assessment and intelligent decision-making, and target detection. This system integrates interference feature learning, dynamic game decision-making and anti-interference execution into a single seeker, breaking through the response bottleneck of traditional discrete architecture, improving the accuracy of interference type identification, enhancing anti-interference capability in complex electromagnetic interference scenarios, increasing the seeker target interception probability in complex electromagnetic countermeasures environments, and significantly enhancing penetration performance.

[0030] Other features and advantages of this application will be described in detail in the following detailed description section. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 A flowchart illustrating the anti-jamming method for a radar seeker based on integrated detection and interception provided in this application embodiment;

[0033] Figure 2A structure schematic diagram of a radar seeker anti-interference system based on a detection and jamming integrated system is provided for an embodiment of the present application.

[0034] Figure 3 A structure schematic diagram of a radar seeker anti-interference device based on a detection and jamming integrated system is provided for an embodiment of the present application Figure 1 ;

[0035] Figure 4 A structure schematic diagram of a radar seeker anti-interference device based on a detection and jamming integrated system is provided for an embodiment of the present application Figure 2 . DETAILED DESCRIPTION

[0036] The technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person skilled in the art without creative work fall within the scope of protection of the present application. Meanwhile, in the description of the embodiments of the present application, the terms "first", "second", etc. are only used for differentiation and description, and cannot be understood as indicating or implying relative importance. Therefore, the features with "first" and "second" can explicitly or implicitly include one or more features. In the description of the embodiments of the present application, the meaning of "multiple" is two or more, unless otherwise specifically limited.

[0037] Figure 1 A flowchart of a radar seeker anti-interference method based on a detection and jamming integrated system is provided for an embodiment of the present application.

[0038] The flow of the radar seeker anti-interference method based on the detection and jamming integrated system is as follows:

[0039] In step S101, a to-be-processed echo signal is acquired, and the to-be-processed echo signal is subjected to feature extraction and fusion to obtain a fusion feature of the to-be-processed echo signal.

[0040] In step S102, the fusion feature of the to-be-processed echo signal is input into an interference recognition model to obtain a category of interference signals in the to-be-processed echo signal.

[0041] In step S103, based on the category of the interference signals in the to-be-processed echo signal, known position information of the radar seeker when the to-be-processed echo signal is collected, and known direction information of the interference source when the to-be-processed echo signal is collected, an interference situation map is generated.

[0042] In step S104, based on the interference situation map, a state space of the radar seeker at any time is obtained.

[0043] Step S105, based on the state space of the radar seeker at any time and the plurality of anti-jamming processing modes in the dynamic strategy library, the reward function is set to determine the reward function value corresponding to each anti-jamming processing mode of the radar seeker, wherein the dynamic strategy library is preset.

[0044] Step S106, based on the reward function value corresponding to each anti-jamming processing mode of the radar seeker, the state space of the radar seeker at any time and the plurality of anti-jamming processing modes in the dynamic strategy library, the expected return corresponding to each anti-jamming processing mode of the radar seeker is calculated by the Q-learning algorithm, and then the anti-jamming processing mode corresponding to the maximum expected return is selected as the optimal anti-jamming processing mode for the echo signal to be processed.

[0045] Step S107, the optimal anti-jamming processing mode is used to process the echo signal to be processed, and the echo signal after anti-jamming processing is obtained, and then the echo signal after anti-jamming processing is processed to detect the target and locate the position of the target.

[0046] In step S101, the echo signal to be processed is generally obtained by wide-band spectrum sensing.

[0047] In addition, the echo signal to be processed is extracted and fused to obtain the fusion feature of the echo signal to be processed. It can be understood that in this step, the echo signal to be processed is first filtered by a low-pass filter to remove high-frequency noise, then amplified by an amplifier, and then converted into a digital signal by an analog-to-digital converter. The digital signal of the echo signal to be processed is obtained; then the multi-dimensional feature extraction and fusion of the digital signal of the echo signal to be processed in time domain, frequency domain and space domain are carried out to obtain the fusion feature of the echo signal to be processed.

[0048] Wherein, the multi-dimensional feature extraction and fusion of the digital signal of the echo signal to be processed in time domain, frequency domain and space domain are carried out to obtain the fusion feature data set, including steps S201-S206, as follows:

[0049] Step S201, the digital signal of the echo signal to be processed is processed by short-time Fourier transform to obtain the time-frequency energy distribution diagram of the echo signal to be processed.

[0050] Step S202, the digital signal of the echo signal to be processed is extracted by time interval to obtain the time sequence pulse sequence of the echo signal to be processed.

[0051] Wherein, the above time sequence pulse sequence includes time domain parameters (pulse width, repetition interval, residence time) of each pulse, sequence length, etc.

[0052] Step S203, spatial domain analysis and extraction are performed on the digital signal of the to-be-processed echo signal to obtain a spatial domain feature vector of the to-be-processed echo signal.

[0053] The spatial domain feature vector includes a wave velocity pointing angle, a polarization parameter, and the like.

[0054] Step S204, a time-frequency energy distribution map of the to-be-processed echo signal is processed by using a CNN to obtain a CNN output of the to-be-processed echo signal.

[0055] The CNN (Convolutional Neural Network) is a convolutional neural network, and when processing a time-frequency map (i.e., a time-frequency energy distribution map), the CNN uses its powerful local feature extraction capability to identify patterns and structures in the time-frequency map.

[0056] In this application, the CNN network architecture is composed of a first layer of convolutional layers, a maximum pooling layer, a second layer of convolutional layers, and a global average pooling layer. The first layer of convolutional layers is composed of 16 3x3 convolutional kernels, which extract the local features of the input time-frequency map (i.e., the time-frequency energy distribution map) through a ReLU activation function. The maximum pooling layer performs a dimensionality reduction operation on the feature map after the first layer of convolution through a 2x2 pooling kernel, retaining the maximum value in each local region, thereby extracting more significant local features and reducing the amount of calculation. The second layer of convolutional layers uses 32 3x3 convolutional kernels to further perform convolution operations on the feature map after the pooling, extracting higher-level and more complex local features, and enhancing the non-linear expression capability through ReLU activation. The global average pooling layer performs global averaging on the output feature map of the second layer of convolutional layers, calculating the average value of each channel to obtain a global feature representation, thereby reducing overfitting and effectively compressing the output.

[0057] Step S205, the time sequence pulse sequence of the to-be-processed echo signal is processed by using an RNN to obtain an RNN output of the to-be-processed echo signal.

[0058] The RNN (Recurrent Neural Network) is a recurrent neural network that can process data with time-dependent relationships through the recurrent connections in the network.

[0059] In this application, the RNN network architecture is composed of a single-layer LSTM layer, which is composed of 64 LSTM units. Each LSTM unit processes the input data of a single time step and controls the flow of information through its internal forget gate, input gate, and output gate, capturing and storing long-term time-dependent information. The recurrent connections between the LSTM units ensure the association between the previous and subsequent time steps, so that the model can maintain the memory of historical information when processing the input time sequence data, thereby outputting a result with time-dependent features.

[0060] In step S206, the CNN output of the echo signal to be processed, the RNN output of the echo signal to be processed and the spatial feature vector of the echo signal to be processed are spliced to obtain the fusion feature of the echo signal to be processed.

[0061] In this application, the CNN output of the echo signal to be processed, the RNN output of the echo signal to be processed and the spatial feature vector of the echo signal to be processed are spliced to obtain the fusion feature of the echo signal to be processed.

[0062] The specific splicing formula is:

[0063] Wherein, The channel dimension splicing (non-element weighted sum) is expressed as, is a weight matrix, n1, n2, n3 are the dimensions of the CNN output, the RNN output and the spatial feature, and d is the dimension of the fusion feature.

[0064] In step S102, the interference identification model is an optimized model, and the training process of the interference identification model includes steps S301-S303, which are as follows:

[0065] In step S301, an echo signal data set is obtained.

[0066] The echo signal data set includes a plurality of echo signals and the true label of the interference signal in each echo signal.

[0067] It should be noted that the echo signal data set is generally obtained by wideband spectrum sensing.

[0068] In step S302, feature extraction and fusion are performed on each echo signal in the echo signal data set to obtain a fusion feature data set.

[0069] The fusion feature data set includes the fusion feature of each echo signal.

[0070] The steps of feature extraction and fusion for each echo signal can refer to the processing of the echo signal to be processed in steps S201-S206 of the present application, and will not be repeated here.

[0071] In step S303, the fusion feature data set is input into the interference identification model, and the interference identification model predicts the class probability of the interference signal in each echo signal based on the fusion feature of each echo signal in the fusion feature data set. Based on the class probability of the interference signal in each echo signal and the true label of the interference signal in each echo signal, the parameters of the interference identification model are optimized using the cross-entropy loss function to obtain the optimized interference identification model.

[0072] It can be understood that the interference identification model's role is to predict the class probability of the interference signal in the echo signal based on the fused features of the echo signal. Furthermore, this interference identification model contains two fully connected layers: the first fully connected layer consists of 128 neurons, employing the ReLU activation function, and is responsible for feature extraction and nonlinear transformation of the input fused feature data; the second fully connected layer serves as the output layer, with the number of neurons matching the number of interference classes, outputting the class probability of the interference signal in each echo signal.

[0073] In this application, it is assumed that there are N echo signals in the echo signal dataset, and there are C categories of interference signals in the echo signal dataset (such as noise interference, deception interference, deceptive false targets, etc.). The fused features corresponding to any echo signal are input into the interference identification model, and the category probability output by the interference identification model is P = [p1, p2, p3, ..., p C The actual labels are y = [y1, y2, y3, ..., y]. C ].

[0074] Based on the class probability of the interfering signal in each echo signal and the true label of the interfering signal in each echo signal, the cross-entropy of all echo signals in the echo signal dataset is obtained using the formula of the cross-entropy loss function. The specific formula is as follows:

[0075]

[0076] Among them, L j Let y be the cross-entropy of the j-th echo signal, C represent the number of categories of interference signals, and y be the cross-entropy of the j-th echo signal. i p represents the true label of the i-th category of the j-th echo signal. i ω represents the predicted probability of the j-th echo signal belonging to the i-th category. i The weight coefficient for the i-th category is inversely proportional to the number of interfering signals in the echo signal dataset that belong to the i-th category.

[0077] Based on the cross-entropy of all echo signals, the parameters of the interference identification model are optimized to obtain the optimized interference identification model.

[0078] It should also be noted that the above ω i The weight coefficient for the i-th category is inversely proportional to the number of interfering signals belonging to the i-th category in the echo signal dataset, which can be specifically expressed as: Where, N i This represents the number of interference signals belonging to the i-th category in the echo signal dataset. λ represents the global scaling factor (default is C = 12), used to balance the weight distribution. ε is a minimum value (1 * 10⁻⁵) to prevent the denominator from being zero.

[0079] As can be seen from the above, since the number of echo signal in the echo signal data set is N, N cross entropy values are obtained, and in the present application, the parameters of the interference identification model can be optimized according to each cross entropy value, in addition, the cross entropy of all echo signals can be summed and divided by the number of echo signals in the echo signal data set to obtain the average cross entropy loss, and the specific formula is:

[0080]

[0081] Wherein, L represents the average cross entropy loss, and N represents the number of echo signals in the echo signal data set.

[0082] Based on the average cross entropy loss, the sample imbalance is solved, the model is prevented from being biased to the majority class, the overfitting is avoided, the sensitivity of the model to noise is reduced, the parameters of the interference identification model are optimized, and the optimized interference identification model is obtained.

[0083] In addition, the above fusion feature data set can also be divided into a training set, a validation set and a test set according to a preset proportion, the model is trained by using the training set by the above method, and the loss function is used for optimization, then the hyperparameters of the interference identification model can be adjusted by using the validation set to avoid overfitting, and the ability of the interference identification model can be evaluated by using the test set, for details, please refer to the prior art, which will not be described here.

[0084] In step S104, based on the interference situation map, the state space of the radar seeker at any time is obtained, which can be understood as:

[0085] Based on the interference situation map, the interference type of the interference signal in the to-be-processed echo signal and the position information of the radar seeker are obtained, and the threat level is further analyzed to obtain the threat level, and the above contents are jointly constructed into the state space of the radar seeker at any time.

[0086] Taking the state space s t at t time as an example, s t = (J type , J threat , S radar ), wherein J type represents interference type code (noise suppression = 1, repeater deception jamming = 2, intermittent sampling jamming = 3, …), a total of 12 types; J threat represents threat level, and S radar represents radar seeker position information (including current frequency, waveform parameter, polarization mode, target tracking error).

[0087] It is to be noted that the multiple anti-interference processing manners in the dynamic strategy library are pre-stored countermeasures matched with different interference types, for example, an adaptive beam forming technology is used to adjust the beam direction of the radar antenna according to the direction of the interference source, so as to reduce the reception of the interference signal; for another example, a frequency agility technology is used to dynamically adjust the radar operating frequency to avoid the interference frequency band; for another example, a sparse reconstruction algorithm is used to reconstruct the target signal to suppress the influence of the interference signal.

[0088] In the examples of the present application, the multiple anti-interference processing manners (i.e. anti-interference actions) are specifically as follows: a1, frequency hopping (FH) + linear frequency modulation (LFM) waveform; a2, polarization switching + phase coding (CPM) waveform; a3, spatial beam zeroing + frequency random jitter, and so on, and there are 8 pre-defined basic actions and 16 combined anti-interference actions.

[0089] By setting the reward function, the reward function values corresponding to the multiple anti-interference processing manners executed by the radar seeker are determined. The reward function is set by considering the tracking error, interference suppression gain and action cost after the radar seeker executes the multiple anti-interference processing manners, and the reward function formula is r t =α·ΔE track +β·ΔJaction suppress , wherein ΔE track represents the tracking error reduction (compared with the previous time, unit: meter); ΔJ suppress represents the interference suppression gain (the signal-to-interference ratio improvement value in dB after space-time-frequency filtering); C action represents the action execution cost (such as frequency switching time, waveform calculation complexity).

[0090] In addition, the weight coefficients in the reward formula are pre-set as α=0.6, β=0.3, and γ=0.1 (dynamically adjusted by policy gradient optimization), and are determined according to the actual situation, which is not limited herein.

[0091] In step S106, based on the reward function values corresponding to the multiple anti-interference processing manners executed by the radar seeker, the state space of the radar seeker at any time and the multiple anti-interference processing manners in the dynamic strategy library, the expected returns corresponding to the multiple anti-interference processing manners executed by the radar seeker are calculated by the Q-learning algorithm, and then the anti-interference processing manner corresponding to the maximum expected return is selected as the optimal anti-interference processing manner for the to-be-processed echo signal.

[0092] Specifically, the expected returns (i.e. Q values) corresponding to the multiple anti-interference processing manners executed by the radar seeker are calculated by the formula:

[0093] , wherein r t ​This is the reward function value, where η represents the learning rate (initial value 0.2, decaying exponentially to 0.05 with training); δ represents the discount factor (0.9, emphasizing long-term returns); Q(s) t ,a t ) represents the state space s t The expected benefit of implementing anti-interference processing method a is as follows: Indicates the next state s t+1 The maximum expected benefit among multiple anti-interference processing methods.

[0094] In step S107, the target detection, identification, and location of the target are performed on the echo signal after anti-jamming processing. This can be understood as:

[0095] To detect, identify, and locate the initial position of the target using the echo signal after anti-jamming processing;

[0096] A multi-target tracking method using particle filtering algorithm is employed to continuously detect targets, estimate their position, velocity, and acceleration in real time, and generate their motion trajectories.

[0097] Furthermore, after step S107, the working parameters of the radar seeker, such as frequency, beam direction, and signal processing algorithm parameters, can be dynamically adjusted based on the target tracking results and the generated target trajectory to further optimize the anti-jamming effect and target detection accuracy.

[0098] In summary, in this embodiment, a closed-loop system of "detection-interference identification-detection" is formed by designing multi-domain feature joint extraction, spatiotemporal-frequency joint filtering, deep learning classification model, interference situation assessment and intelligent decision-making, and target detection. This integrates interference feature learning, dynamic game decision-making, and anti-interference execution into a single seeker, breaking through the response bottleneck of traditional discrete architectures, improving the accuracy of interference type identification, enhancing anti-interference capabilities in complex electromagnetic interference scenarios, increasing the seeker target interception probability in complex electromagnetic countermeasures environments, and significantly enhancing penetration performance.

[0099] The above combination Figure 1 This application provides a detailed description of the anti-jamming method for a radar seeker based on integrated detection and interception, as described below. Figure 2 This document provides a detailed description of the radar seeker anti-jamming system based on integrated detection and interception provided in the embodiments of this application.

[0100] The system specifically includes: an acquisition module, a feature extraction and fusion module, a model processing module, an interference situation assessment and intelligent decision-making module, and a detection and tracking module, as shown below.

[0101] The acquisition module is used to acquire the echo signal to be processed;

[0102] The feature extraction and fusion module is configured to perform feature extraction and fusion on the echo signal to be processed to obtain fused features of the echo signal to be processed.

[0103] The model processing module is configured to input the fused features of the echo signal to be processed into an interference recognition model to obtain a category of the interference signal in the echo signal to be processed.

[0104] The interference situation assessment and intelligent decision module is configured to generate an interference situation graph based on the category of the interference signal in the echo signal to be processed, known position information of the radar seeker at the time of collection of the echo signal to be processed, and known direction information of the interference source at the time of collection of the echo signal to be processed, and obtain a state space of the radar seeker at any time based on the interference situation graph. The interference situation assessment and intelligent decision module is further configured to determine reward function values corresponding to the radar seeker performing a plurality of anti-interference processing modes respectively by setting a reward function based on the state space of the radar seeker at any time and the plurality of anti-interference processing modes in a dynamic strategy library. The interference situation assessment and intelligent decision module is further configured to calculate expected returns corresponding to the radar seeker performing the plurality of anti-interference processing modes respectively by a Q-learning algorithm based on the reward function values corresponding to the radar seeker performing the plurality of anti-interference processing modes respectively, the state space of the radar seeker at any time, and the plurality of anti-interference processing modes in the dynamic strategy library, and select an anti-interference processing mode corresponding to a maximum expected return as an optimal anti-interference processing mode for the echo signal to be processed. The dynamic strategy library is preset.

[0105] The detection and tracking module is configured to perform anti-interference processing on the echo signal to be processed by using the optimal anti-interference processing mode to obtain an echo signal processed by anti-interference, and perform target detection on the echo signal processed by anti-interference to identify and locate a position of the target.

[0106] In addition, for the above system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the part of the description of the method embodiments. Moreover, it should be noted that in each module of the system of the present application, the components therein are logically divided according to the functions to be implemented, but the present application is not limited thereto, and each component can be re-divided or combined as needed.

[0107] The above describes the radar seeker anti-interference method and system based on the integration of detection, interference and tracking provided by the embodiments of the present application. The following will be described in detail Figures 3-4 The radar seeker anti-interference device based on the integration of detection, interference and tracking provided by the embodiments of the present application is described in detail.

[0108] Figure 3 is a structural schematic diagram of the radar seeker anti-interference device based on the integration of detection, interference and tracking provided by the embodiments of the present application Figure 1 . For example, Figure 3As shown, the radar seeker anti-jamming device 300 based on integrated detection and interception includes: a transceiver module 301 and a processing module 302. For ease of explanation, Figure 3 Only the main components of this radar seeker anti-jamming device based on integrated detection and interception are shown.

[0109] The transceiver module 301 is used to perform the transceiver function of the above-mentioned radar seeker anti-jamming method based on integrated detection and interference, and the processing module 302 is used to perform other functions of the above-mentioned radar seeker anti-jamming method based on integrated detection and interference.

[0110] Optionally, the transceiver module 301 may include a sending module ( Figure 3 (not shown in the image) and receiving module ( Figure 3 (Not shown in the image). The transmitting module is used to implement the transmitting function of the radar seeker anti-jamming device 300 based on integrated detection and interception, and the receiving module is used to implement the receiving function of the radar seeker anti-jamming device 300 based on integrated detection and interception.

[0111] Optionally, the radar seeker anti-jamming device 300 based on integrated detection and interception may also include a storage module. Figure 3 (Not shown in the image), the storage module stores programs or instructions. When the processing module 302 executes the program or instructions, the radar seeker anti-jamming device 300 based on integrated detection and interception can perform the radar seeker anti-jamming method based on integrated detection and interception in the embodiments of this application.

[0112] The following is combined with Figure 4 A detailed description of each component of the radar seeker anti-jamming device 400 based on integrated detection and interception is provided below:

[0113] The processor 401 is the control center of the radar seeker anti-jamming device 400 based on the integrated detection and interception system. It can be a single processor or a collective term for multiple processing elements. For example, the processor 401 can be one or more central processing units (CPUs), application-specific integrated circuits (ASICs), or one or more integrated circuits configured to implement the embodiments of this application, such as one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs).

[0114] Optionally, the processor 401 can execute various functions of the radar seeker anti-jamming device 400 based on the integration of the detection and the jamming by running or executing software programs stored in the memory 402 and calling data stored in the memory 402, such as executing the radar seeker anti-jamming method based on the integration of the detection and the jamming in the embodiments of the present application.

[0115] In a specific implementation, as an embodiment, the processor 401 can include one or more CPUs, such as the CPU0 and the CPU1 shown in FIG. 4. Figure 4

[0116] In a specific implementation, as an embodiment, the radar seeker anti-jamming device 400 based on the integration of the detection and the jamming can also include multiple processors, such as the processor 401 and the processor 404 shown in FIG. 4. Figure 4

[0117] Optionally, the memory 402 can be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disk storage, a magneto-optical disk, a magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited to this. The memory 402 can be integrated with the processor 401 or exist independently and be coupled to the processor 401 through an interface circuit (not shown) of the radar seeker anti-jamming device 400 based on the integration of the detection and the jamming, and the embodiments of the present application are not limited in this regard. Figure 4

[0118] ​​​The transceiver 403 is used for communication with other communication devices. For example, in a radar seeker anti-jamming device 400 that integrates detection and interception, which is the first device, the transceiver 403 can be used to communicate with a second device or a third device.

[0119] Alternatively, transceiver 403 may include a receiver and a transmitter. Figure 4 (Not shown separately). The receiver is used to implement the receiving function, and the transmitter is used to implement the sending function.

[0120] Optionally, the transceiver 403 can be integrated with the processor 401, or it can exist independently, and can be connected via the interface circuit of the radar seeker anti-jamming device 400 based on the integrated detection and interference detection system. Figure 4 (Not shown in the image) is coupled to processor 401, and this embodiment of the application does not specifically limit this.

[0121] Understandable Figure 4 The structure of the radar seeker anti-jamming device 400 based on integrated detection and interference shown in the figure does not constitute a limitation on the radar seeker anti-jamming device based on integrated detection and interference. The actual radar seeker anti-jamming device based on integrated detection and interference may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0122] Furthermore, the technical effects of the radar seeker anti-jamming device 400 based on the integrated detection and interception system can be referred to the technical effects of the method described in the above method embodiments, and will not be repeated here.

[0123] It should be understood that the processor in the embodiments of this application can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0124] It should also be understood that the memory in the embodiments of the present application can be a volatile memory or a nonvolatile memory, or can include both volatile and nonvolatile memory. Among them, the nonvolatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically EPROM (EEPROM), or a flash memory. The volatile memory can be a random access memory (RAM) used as an external cache. By way of example, and not limitation, many forms of random access memory (RAM) are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0125] The above-described embodiments can be implemented in whole or in part by software, hardware (such as a circuit), firmware, or any combination thereof. When implemented in software, the above-described embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are wholly or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through wired (such as infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. containing one or more available medium sets. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid state disk.

Claims

1. A radar seeker anti-jamming method based on integrated detection and tracking, characterized in that, The method comprises: acquiring a to-be-processed echo signal, and performing feature extraction and fusion on the to-be-processed echo signal to obtain fused features of the to-be-processed echo signal; inputting the fused features of the to-be-processed echo signal into an interference identification model to obtain a category of interference signals in the to-be-processed echo signal; generating an interference situation map based on the category of interference signals in the to-be-processed echo signal, known position information of a radar seeker at a time when the to-be-processed echo signal is collected, and known direction information of an interference source at the time when the to-be-processed echo signal is collected; obtaining a state space of the radar seeker at any time based on the interference situation map; based on the state space of the radar seeker at any time and a plurality of anti-interference processing modes in a dynamic strategy library, determining reward function values corresponding to the plurality of anti-interference processing modes by setting a reward function, wherein the dynamic strategy library is preset; based on the reward function values corresponding to the plurality of anti-interference processing modes, the state space of the radar seeker at any time, and the plurality of anti-interference processing modes in the dynamic strategy library, calculating expected returns corresponding to the plurality of anti-interference processing modes of the radar seeker by a Q-learning algorithm, and then selecting an anti-interference processing mode corresponding to a maximum expected return as an optimal anti-interference processing mode for the to-be-processed echo signal; performing anti-interference processing on the to-be-processed echo signal by using the optimal anti-interference processing mode to obtain an anti-interference processed echo signal, and then performing target detection on the anti-interference processed echo signal to identify and locate a position of a target.

2. The radar seeker anti-jamming method based on the integration of the detection and the jamming according to claim 1, characterized in that, The method comprises: performing feature extraction and fusion on the to-be-processed echo signal to obtain fused features of the to-be-processed echo signal, comprising: removing high-frequency noise from the to-be-processed echo signal through a low-pass filter, amplifying the to-be-processed echo signal through an amplifier, and converting an analog signal into a digital signal through an analog-to-digital converter to obtain a digital signal of the to-be-processed echo signal; 3. The radar seeker anti-jamming method based on the integration of the detection and the jamming according to claim 2, characterized in that, performing multi-dimensional feature extraction and fusion on the digital signal of the to-be-processed echo signal in a time domain, a frequency domain, and a space domain to obtain the fused features of the to-be-processed echo signal. The method comprises: performing short-time Fourier transform processing on the digital signal of the to-be-processed echo signal to obtain a time-frequency energy distribution map of the to-be-processed echo signal; extracting pulses of the digital signal of the to-be-processed echo signal according to time intervals to obtain a time sequence pulse sequence of the to-be-processed echo signal; performing spatial domain analysis and extraction on the digital signal of the to-be-processed echo signal to obtain a space domain feature vector of the to-be-processed echo signal; processing the time-frequency energy distribution map of the to-be-processed echo signal by using a CNN to obtain a CNN output of the to-be-processed echo signal; processing the time sequence pulse sequence of the to-be-processed echo signal by using an RNN to obtain an RNN output of the to-be-processed echo signal; Concatenate the CNN output of the to-be-processed echo signal, the RNN output of the to-be-processed echo signal, and the spatial feature vector of the to-be-processed echo signal to obtain a fusion feature of the to-be-processed echo signal.

4. The radar seeker anti-jamming method based on the integration of the detection and the jamming according to claim 1, characterized in that, The interference identification model is an optimized model, and a training step of the interference identification model comprises: obtaining an echo signal dataset, wherein the echo signal dataset comprises a plurality of echo signals and a true label of an interference signal in each echo signal; extracting and fusing features of each echo signal in the echo signal dataset to obtain a fusion feature dataset, wherein the fusion feature dataset comprises a fusion feature of each echo signal; inputting the fusion feature dataset into the interference identification model, and predicting a class probability of the interference signal in each echo signal based on the fusion feature of each echo signal in the fusion feature dataset; and 5. The radar seeker anti-jamming method based on the integration of the jamming detection according to claim 4, characterized in that, optimizing parameters of the interference identification model based on the class probability of the interference signal in each echo signal and the true label of the interference signal in each echo signal, and obtaining an optimized interference identification model by using a cross-entropy loss function. The method comprises: wherein the L j represents the cross entropy of the jth echo signal, the C represents the number of categories of interference signals, the y i represents the true label of the ith category of the jth echo signal, the p i represents the predicted probability of the ith category of the jth echo signal, the ω i is inversely proportional to the number of interference signals belonging to the ith category in the echo signal in the echo signal data set. obtaining an echo signal dataset, wherein the echo signal dataset comprises a plurality of echo signals and a true label of an interference signal in each echo signal; 6. The radar seeker anti-jamming method based on the integration of the detection and the jamming according to claim 5, characterized in that, extracting and fusing features of each echo signal in the echo signal dataset to obtain a fusion feature dataset, wherein the fusion feature dataset comprises a fusion feature of each echo signal; inputting the fusion feature dataset into the interference identification model, and predicting a class probability of the interference signal in each echo signal based on the fusion feature of each echo signal in the fusion feature dataset; and optimizing parameters of the interference identification model based on the class probability of the interference signal in each echo signal and the true label of the interference signal in each echo signal, and obtaining an optimized interference identification model by using a cross-entropy loss function. The method comprises:

7. The radar seeker anti-jamming method based on the integration of the jammer detection according to claim 1, characterized in that, obtaining an echo signal dataset, wherein the echo signal dataset comprises a plurality of echo signals and a true label of an interference signal in each echo signal; extracting and fusing features of each echo signal in the echo signal dataset to obtain a fusion feature dataset, wherein the fusion feature dataset comprises a fusion feature of each echo signal; inputting the fusion feature dataset into the interference identification model, and predicting a class probability of the interference signal in each echo signal based on the fusion feature of each echo signal in the fusion feature dataset; and 8. A radar seeker anti-jamming system based on the integration of the detection and the jamming, characterized in that, optimizing parameters of the interference identification model based on the class probability of the interference signal in each echo signal and the true label of the interference signal in each echo signal, and obtaining an optimized interference identification model by using a cross-entropy loss function. The method comprises: obtaining an echo signal dataset, wherein the echo signal dataset comprises a plurality of echo signals and a true label of an interference signal in each echo signal; extracting and fusing features of each echo signal in the echo signal dataset to obtain a fusion feature dataset, wherein the fusion feature dataset comprises a fusion feature of each echo signal; inputting the fusion feature dataset into the interference identification model, and predicting a class probability of the interference signal in each echo signal based on the fusion feature of each echo signal in the fusion feature dataset; and optimizing parameters of the interference identification model based on the class probability of the interference signal in each echo signal and the true label of the interference signal in each echo signal, and obtaining an optimized interference identification model by using a cross-entropy loss function. The method comprises: obtaining an echo signal dataset, wherein the echo signal dataset comprises a plurality of echo signals and a true label of an interference signal in each echo signal; extracting and fusing features of each echo signal in the echo signal dataset to obtain a fusion feature dataset, wherein the fusion feature dataset comprises a fusion feature of each echo signal; inputting the fusion feature dataset into the interference identification model, and predicting a class probability of the interference signal in each echo signal based on the fusion feature of each echo signal in the fusion feature dataset; and optimizing parameters of the interference identification model based on the class probability of the interference signal in each echo signal and the true label of the interference signal in each echo signal, and obtaining an optimized interference identification model by using a cross-entropy loss function. The method comprises: obtaining an echo signal dataset, wherein the echo signal dataset comprises a plurality of echo signals and a true label of an interference signal in each echo signal; extracting and fusing features of each echo signal in the echo signal dataset to obtain a fusion feature dataset, wherein the fusion feature dataset comprises a fusion feature of each echo signal; inputting the fusion feature dataset into the interference identification model, and predicting a class probability of the interference signal in each echo signal based on the fusion feature of each echo signal in the fusion feature dataset; and optimizing parameters of the interference identification model based on the class probability of the interference signal in each echo signal and the true label of the interference signal in each echo signal, and obtaining an optimized interference identification model by using a cross-entropy loss function. The method comprises: obtaining an echo signal dataset, wherein the echo signal dataset comprises a plurality of echo signals and a true label of an interference signal in each echo signal; extracting and fusing features of each echo signal in the echo signal dataset to obtain a fusion feature dataset, wherein the fusion feature dataset comprises a fusion feature of each echo signal; inputting the fusion feature dataset into the interference identification model, and predicting a class probability of the interference signal in each echo signal based on the fusion feature of each echo signal in the fusion feature dataset; and optimizing parameters of the interference identification model based on the class probability of the interference signal in each echo signal and the true label of the interference signal in each echo signal, and obtaining an optimized interference identification model by using a cross-entropy loss function. The system comprises: an acquisition module configured to acquire a to-be-processed echo signal; a feature extraction and fusion module configured to extract and fuse features of the to-be-processed echo signal to obtain a fusion feature of the to-be-processed echo signal; a model processing module configured to input the fusion feature of the to-be-processed echo signal into an interference identification model to obtain a class of an interference signal in the to-be-processed echo signal; and The interference situation assessment and intelligent decision module is configured to generate an interference situation graph based on the category of the interference signal in the to-be-processed echo signal, known position information of the radar seeker when the to-be-processed echo signal is collected, and known direction information of the interference source when the to-be-processed echo signal is collected, and obtain a state space of the radar seeker at any time based on the interference situation graph; the radar seeker is configured to determine reward function values corresponding to the radar seeker executing a plurality of anti-interference processing manners based on the state space of the radar seeker at any time and the plurality of anti-interference processing manners in a dynamic strategy library by setting a reward function; the radar seeker is further configured to calculate expected returns of the radar seeker executing the plurality of anti-interference processing manners corresponding respectively by a Q-learning algorithm based on the reward function values corresponding respectively to the radar seeker executing the plurality of anti-interference processing manners, the state space of the radar seeker at any time, and the plurality of anti-interference processing manners in the dynamic strategy library, and select an anti-interference processing manner corresponding to a maximum expected return as an optimal anti-interference processing manner for the to-be-processed echo signal, wherein the dynamic strategy library is preset; The detection and tracking module is configured to perform anti-interference processing on the to-be-processed echo signal by using the optimal anti-interference processing manner to obtain an anti-interference processed echo signal, and perform target detection on the anti-interference processed echo signal to identify and locate a position of a target.

9. A radar seeker anti-jamming device based on the integration of the detection and the jamming, characterized in that, The apparatus includes means for performing the method of any one of claims 1-7.

Images