Radar working mode recognition method and device, equipment, medium and program product

Abstract

The invention provides a radar working mode recognition method, device and equipment, a medium and a program product, and relates to the technical field of artificial intelligence. The method comprises the following steps: acquiring a pulse description word (PDW) sequence of a radar radiation source; based on the PDW sequence, determining intra-pulse features of the radar radiation source and inter-pulse features of the radar radiation source; inputting the intra-pulse feature and the inter-pulse feature into a working mode feature extraction model to obtain an intra-pulse feature vector and an inter-pulse feature vector output by the working mode feature extraction model; and inputting a feature fusion vector of the intra-pulse feature vector and the inter-pulse feature vector into a radar working mode recognition model to obtain a radar working mode recognition result output by the radar working mode recognition model. According to the method, the intra-pulse features and the inter-pulse features are obtained firstly, then the intra-pulse features and the inter-pulse features are subjected to feature extraction, the intra-pulse feature vectors and the inter-pulse feature vectors with context information are obtained, and therefore high-precision recognition of the radar working mode can be achieved for all scenes.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a radar operating mode recognition method, device, equipment, medium and program product. Background Art

[0002] Radar operating modes vary the radar's signal transmission pattern and parameters to adapt to different targets, interference environments, and missions. These modes include, but are not limited to, velocity search, track-and-search, ranging-and-track, multi-target tracking, tracking-and-search, and single-target tracking. Radar operating mode identification is often required in radar applications.

[0003] Currently, radar operating mode recognition is accomplished by statistically analyzing and storing pulse descriptors such as carrier frequency, pulse width, and pulse repetition interval of different radar emitters. This data is then matched against the radar operating modes in the template library. However, this existing technology is only suitable for scenarios with large variations in radar signal parameters and a uniform electromagnetic environment. For other scenarios, this radar operating mode recognition solution has low accuracy. Summary of the Invention

[0004] The present invention provides a radar working mode recognition method, device, equipment, medium and program product, which are used to solve the defect of low accuracy of radar working mode recognition in the prior art and realize high-accuracy radar working mode recognition.

[0005] The present invention provides a radar operating mode recognition method, comprising: Acquire a pulse descriptor word PDW sequence of a radar radiation source; the PDW sequence includes PDWs at multiple moments; determining, based on the PDW sequence, intra-pulse characteristics of the radar emitter and inter-pulse characteristics of the radar emitter; Inputting the intra-pulse feature and the inter-pulse feature into a working mode feature extraction model to obtain an intra-pulse feature vector and an inter-pulse feature vector output by the working mode feature extraction model; the working mode feature extraction model is used to extract an intra-pulse feature vector with context information based on the intra-pulse feature, and the working mode feature extraction model is further used to extract an inter-pulse feature vector with context information based on the inter-pulse feature; The feature fusion vector of the intra-pulse feature vector and the inter-pulse feature vector is input into a radar working mode recognition model to obtain a radar working mode recognition result output by the radar working mode recognition model; the radar working mode recognition model is used to perform radar working mode recognition on the feature fusion vector.

[0006] According to a radar operating mode recognition method provided by the present invention, any of the PDWs includes carrier frequency, pulse width and arrival time; The determining, based on the PDW sequence, the intra-pulse characteristics and the inter-pulse characteristics of the radar emitter includes: Extracting the carrier frequency, pulse width and arrival time of the PDW sequence respectively to obtain a carrier frequency sequence, a pulse width sequence and an arrival time sequence; splicing the carrier frequency sequence and the pulse width sequence to obtain the intra-pulse feature; The arrival time series is subjected to differential processing to obtain the inter-pulse features.

[0007] According to a radar operating mode recognition method provided by the present invention, the step of inputting the intra-pulse feature and the inter-pulse feature into an operating mode feature extraction model to obtain an intra-pulse feature vector and an inter-pulse feature vector output by the operating mode feature extraction model includes: Inputting the intra-pulse feature into the first long short-term memory network layer in the working mode feature extraction model to obtain the intra-pulse feature vector output by the first long short-term memory network layer; Inputting the inter-pulse feature into a second long short-term memory network layer in the working mode feature extraction model to obtain an inter-pulse feature vector output by the second long short-term memory network layer; The working mode feature extraction model is obtained by performing unsupervised training based on the sample PDW sequence.

[0008] According to a radar working mode recognition method provided by the present invention, the working mode feature extraction model is a two-way contrast prediction coding model, and the working mode feature extraction model is used to extract essential features.

[0009] According to a radar operating mode recognition method provided by the present invention, after obtaining the pulse descriptor word PDW sequence of the radar radiation source, the method further includes: Inputting the PDW sequence into a radar operating mode detection model to obtain a radar operating mode detection result output by the radar operating mode detection model; the radar operating mode detection model is used to perform radar operating mode detection based on the PDW sequence; After inputting the feature fusion vector of the intra-pulse feature vector and the inter-pulse feature vector into the radar working mode recognition model to obtain the radar working mode recognition result output by the radar working mode recognition model, the method further includes: Based on the radar working mode recognition result and the radar working mode detection result, a final radar working mode recognition result is determined.

[0010] According to a radar operating mode recognition method provided by the present invention, inputting the PDW sequence into a radar operating mode detection model to obtain a radar operating mode detection result output by the radar operating mode detection model includes: Inputting each PDW in the PDW sequence into a radar working mode detection model, respectively, to obtain a radar working mode detection result output by the radar working mode detection model; The radar operating mode detection model is used to perform radar operating mode detection based on one PDW, and the radar operating mode detection result includes radar operating mode detection sub-results corresponding to each PDW.

[0011] The present invention also provides a radar operating mode recognition device, comprising: A sequence acquisition module is used to acquire a pulse description word PDW sequence of a radar emitter; the PDW sequence includes PDWs at multiple moments; a feature determination module, configured to determine an intra-pulse feature and an inter-pulse feature of the radar emitter based on the PDW sequence; a feature extraction module, configured to input the intra-pulse feature and the inter-pulse feature into a working mode feature extraction model to obtain an intra-pulse feature vector and an inter-pulse feature vector output by the working mode feature extraction model; the working mode feature extraction model is configured to extract an intra-pulse feature vector having contextual information based on the intra-pulse feature, and the working mode feature extraction model is further configured to extract an inter-pulse feature vector having contextual information based on the inter-pulse feature; The pattern recognition module is used to input the feature fusion vector of the intra-pulse feature vector and the inter-pulse feature vector into the radar working mode recognition model to obtain the radar working mode recognition result output by the radar working mode recognition model; the radar working mode recognition model is used to perform radar working mode recognition on the feature fusion vector.

[0012] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the radar operating mode recognition method described above is implemented.

[0013] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the radar operating mode recognition method described in any one of the above is implemented.

[0014] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements any of the above-mentioned radar operating mode recognition methods.

[0015] The radar working mode recognition method, apparatus, equipment, medium and program product provided by the present invention obtain the PDW sequence of the radar radiation source, and the PDW sequence includes PDWs at multiple moments, so that even if the PDW at some moments includes erroneous parameters, it can be accurately identified based on the PDWs at multiple moments, thereby improving the recognition accuracy of the radar working mode; based on the PDW sequence, the intra-pulse characteristics of the radar radiation source and the inter-pulse characteristics of the radar radiation source are determined, so that the radar working mode is accurately identified based on the intra-pulse characteristics and the inter-pulse characteristics, thereby improving the recognition accuracy of the radar working mode; the intra-pulse characteristics and the inter-pulse characteristics are input into the working mode feature extraction model to obtain the intra-pulse feature vector and the inter-pulse feature vector output by the working mode feature extraction model, based on which, further feature extraction is performed on the intra-pulse characteristics and the inter-pulse characteristics respectively, and working mode feature extraction can be performed on the parameters inside the radar pulse and between the radar pulses respectively, thereby ensuring the extraction of more comprehensive working mode features, thereby improving the recognition of the radar working mode. Accuracy; and the working mode feature extraction model is used to extract the intra-pulse feature vector with context information based on the intra-pulse feature, and the working mode feature extraction model is also used to extract the inter-pulse feature vector with context information based on the inter-pulse feature, so as to fully extract the implicit features and time series correlation information within the PDW sequence, thereby improving the recognition accuracy of the radar working mode; based on the above-mentioned determined input features and feature extraction method, the feature fusion vector of the intra-pulse feature vector and the inter-pulse feature vector is input into the radar working mode recognition model, and an accurate radar working mode recognition result can be obtained. At the same time, the intra-pulse feature and the inter-pulse feature are first obtained based on the PDW sequence, and then the working mode feature extraction is performed on the intra-pulse feature and the inter-pulse feature to obtain the intra-pulse feature vector and the inter-pulse feature vector with context information, so that for complex electromagnetic environment scenarios such as incomplete radar signal parameters or overlapping radar signal parameters, high-precision recognition of the radar working mode can be achieved, and high-precision recognition can be achieved regardless of the scenario. In summary, the present invention can achieve high-precision recognition of the radar working mode for each scenario, that is, the present invention can improve the recognition accuracy of the radar working mode and enhance the adaptability of the radar working mode. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0017] Figure 1 This is one of the flow charts of the radar working mode recognition method provided by the present invention.

[0018] Figure 2This is the second flow chart of the radar working mode recognition method provided by the present invention.

[0019] Figure 3 This is the third flow chart of the radar working mode recognition method provided by the present invention.

[0020] Figure 4 This is the fourth flow chart of the radar working mode recognition method provided by the present invention.

[0021] Figure 5 This is the fifth flow chart of the radar working mode recognition method provided by the present invention.

[0022] Figure 6 This is the sixth flow chart of the radar working mode recognition method provided by the present invention.

[0023] Figure 7 It is a structural diagram of the radar working mode recognition device provided by the present invention.

[0024] Figure 8 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0025] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0026] Current radar mode recognition solutions mostly rely on template matching, which relies on expert experience. Specifically, radar modes are matched one by one against a library of radar modes. However, template matching is only suitable for scenarios with large variations in radar signal parameters and a uniform electromagnetic environment. In other scenarios, this template matching method has low accuracy.

[0027] Given the low accuracy of current radar operating mode recognition schemes, the present inventors conducted research and discovered that existing radar operating mode recognition schemes primarily use several dimensional parameters of radar pulse descriptors, extract statistical features through mathematical statistics, and then perform template matching. However, these features only calculate the distribution patterns of pulse descriptors, resulting in low accuracy in template matching. Based on this, the present inventors initially proposed a pattern recognition method based on traditional machine learning, namely, directly inputting several dimensional parameters of radar pulse descriptors into a machine learning multi-classifier.

[0028] The present inventors studied the above-mentioned approach and found that, while it does not require template matching, it still primarily uses several dimensional parameters of radar pulse descriptors and extracts statistical features through mathematical statistical methods, failing to extract deeper grammatical patterns and temporal correlations. Therefore, the feature description information of this approach is incomplete and incomplete. While it is effective for radar operating mode recognition in fixed scenarios, it is less adaptable to complex electromagnetic environments or scenarios with overlapping radar signal parameters. In other words, it is only suitable for scenarios with large differences in radar signal parameters and a single electromagnetic environment. For other scenarios, the radar operating mode recognition accuracy of this approach remains low.

[0029] To address the challenges presented by the aforementioned approach, the present inventors continued their research and developed the concept of artificially designing physical features, such as signal envelope, amplitude, and syntactic features. These features were constructed based on pulse descriptors and then fed into multiple classifiers for radar mode recognition. However, while these artificially designed features offered good interpretability in terms of physical meaning, their information expression was limited, and some implicit features within the radar descriptor sequences were not fully exploited.

[0030] In response to the defects of the above-mentioned ideas, the present invention further studies and proposes a radar working mode recognition method. The radar working mode recognition method extracts intra-pulse feature vectors with context information based on intra-pulse features, and extracts inter-pulse feature vectors with context information based on inter-pulse features, thereby mining the implicit features and temporal correlation information within the radar description word sequence, thereby improving the accuracy of radar working mode recognition.

[0031] Next, the radar working mode recognition method provided by the present invention is introduced through the following embodiments. Figures 1-6 The radar working mode recognition method of the present invention is described.

[0032] Figure 1 This is one of the flow charts of the radar working mode recognition method provided by the present invention, such as Figure 1 As shown, the radar working mode recognition method includes the following steps 110, 120, 130 and 140.

[0033] Step 110: Acquire a pulse descriptor word (PDW) sequence of a radar emitter.

[0034] Here, the radar radiation source is a device in a radar system that is responsible for generating, modulating, and transmitting electromagnetic wave signals, and the radar radiation source is the radiation source to be identified by the radar operating mode. In one embodiment, the radar radiation source includes a transmitter, an antenna system, and a frequency synthesizer.

[0035] Here, the PDW (Pulse Description Word) sequence includes PDWs at multiple time instants, and this PDW sequence is the sequence to be used for radar operating mode identification. A PDW is a digital description of key parameters of a radar pulse signal from a radar emitter, which can be used to identify and classify the radar emitter. In one embodiment, the PDW is obtained by extracting parameters from the radar pulse signal after receiving it.

[0036] The PDW may include, but is not limited to, at least one of the following: carrier frequency (CF), pulse width (PW), time of arrival (TOA), direction of arrival (DOA), pulse amplitude (PA), pulse repetition interval (PRI), etc.

[0037] It's important to note that the PDW sequence is acquired because it includes PDWs at multiple moments. This allows for the subsequent extraction of contextual features, specifically the mining of implicit features and temporal correlations within the PDW sequence, thereby improving the accuracy of radar mode recognition. Furthermore, the PDWs in the PDW sequence are relatively short, so acquiring a PDW sequence for radar mode recognition improves accuracy compared to using only a single PDW sequence. This prevents erroneous recognition caused by erroneous parameters in the PDWs at certain moments.

[0038] Step 120: Determine the intra-pulse characteristics and the inter-pulse characteristics of the radar emitter based on the PDW sequence.

[0039] Intrapulse features are used to characterize the signal characteristics and parameters within a single radar pulse. These intrapulse features can be determined using at least one parameter in the PDW sequence. The specific determination method can refer to existing techniques or the method studied in this invention. The method studied in this invention involves extracting the carrier frequency and pulse width of the PDW sequence separately to obtain a carrier frequency sequence and a pulse width sequence, then concatenating the carrier frequency sequence and pulse width sequence to obtain the intrapulse features.

[0040] Here, the inter-pulse feature characterizes the parameter variation patterns between different radar pulses. This inter-pulse feature can be determined using at least one parameter in the PDW sequence. The specific determination method can refer to existing techniques or the method studied in this invention. The method studied in this invention involves extracting the arrival times of the PDW sequence to obtain an arrival time series, and performing differential processing on the arrival time series to obtain the inter-pulse feature.

[0041] It should be understood that extracting intra-pulse features can improve the recognition accuracy of the radar operating mode, and extracting inter-pulse features can also improve the recognition accuracy of the radar operating mode.

[0042] Step 130: Input the intra-pulse feature and the inter-pulse feature into a working mode feature extraction model to obtain an intra-pulse feature vector and an inter-pulse feature vector output by the working mode feature extraction model.

[0043] Here, the working mode feature extraction model is used to extract a feature that can effectively characterize the properties of the radar working mode, thereby facilitating the subsequent accurate identification of the radar working mode.

[0044] The working mode feature extraction model is a deep learning model. For example, the working mode feature extraction model may include but is not limited to at least one of the following: a recurrent neural network (RNN), a long short-term memory network (LSTM), a convolutional neural network (CNN), a residual neural network (ResNet), a generative adversarial neural network (GAN), or a transformer network, etc., thereby utilizing the superior expression ability of the deep learning network to extract deep patterns, that is, to extract implicit features within the PDW sequence, thereby improving the accuracy of radar working mode recognition.

[0045] In one embodiment, the working mode feature extraction model is obtained by supervised training based on the sample PDW sequence, and the specific training method can be trained using existing technologies. In another embodiment, the working mode feature extraction model is obtained by unsupervised training based on the sample PDW sequence, and the training method can be a comparative learning method.

[0046] The working mode feature extraction model is configured to extract an intra-pulse feature vector containing contextual information based on the intra-pulse feature, and the working mode feature extraction model is further configured to extract an inter-pulse feature vector containing contextual information based on the inter-pulse feature. That is, the working mode feature extraction model is configured to further extract working mode features containing contextual information.

[0047] In one embodiment, the working mode feature extraction model includes a long short-term memory network layer, which is used to extract working mode features with context information.

[0048] It should be understood that extracting inter-pulse feature vectors and intra-pulse feature vectors with contextual information can fully extract (mine) the implicit features and temporal correlation information within the PDW sequence, that is, extracting deeper grammatical rules and temporal correlation relationships, that is, extracting more comprehensive and complete features, thereby improving the accuracy of radar working mode recognition.

[0049] In one embodiment, the working mode feature extraction model includes two branches, such as a first working mode feature extraction layer and a second working mode feature extraction layer. Specifically, intra-pulse features are input to the first working mode feature extraction layer to obtain an intra-pulse feature vector output by the first working mode feature extraction layer; inter-pulse features are input to the second working mode feature extraction layer to obtain an inter-pulse feature vector output by the second working mode feature extraction layer.

[0050] It should be understood that further feature extraction of intra-pulse features and inter-pulse features can be performed to extract contextual information from features within and between radar pulses, respectively, thereby ensuring the extraction of more comprehensive operating mode features and thereby improving the accuracy of radar operating mode recognition. It should be noted that, through continuous research, the present invention has discovered that further feature extraction of intra-pulse features and inter-pulse features significantly improves the accuracy of radar operating mode recognition.

[0051] Step 140: Input the feature fusion vector of the intra-pulse feature vector and the inter-pulse feature vector into a radar working mode recognition model to obtain a radar working mode recognition result output by the radar working mode recognition model.

[0052] Here, the feature fusion vector is obtained by fusing the intra-pulse feature vector and the inter-pulse feature vector. This feature fusion method can include, but is not limited to, concatenation, element-by-element operations (element-by-element addition, multiplication, or averaging of vectors of the same dimension), weighted fusion, and attention mechanism fusion.

[0053] The radar working mode recognition model is used to perform radar working mode recognition on the feature fusion vector.

[0054] In one specific embodiment, the radar operating mode recognition model is a classification model, eliminating the need for template matching and improving the accuracy of radar operating mode recognition. Furthermore, the radar operating mode recognition model includes a deep residual network (ResNet), which, through residual learning, increases the depth of the network model without causing network degradation, thereby effectively improving the accuracy of radar operating mode recognition.

[0055] In one embodiment, the radar operating mode recognition model is obtained by supervised training based on sample PDW sequences, and the specific training method can be trained using existing technologies. In another embodiment, the radar operating mode recognition model is obtained by unsupervised training based on sample PDW sequences, and the specific training method can be trained using existing technologies.

[0056] Here, the radar working mode recognition result is used to indicate the radar working mode of the radar radiation source. The radar working mode may include but is not limited to: speed search, tracking plus search, tracking while ranging, multi-target tracking, tracking while searching, and single target tracking, etc.

[0057] The radar working mode recognition method provided by the embodiment of the present invention obtains the PDW sequence of the radar radiation source, and the PDW sequence includes PDWs at multiple moments, so that even if the PDW at some moments includes erroneous parameters, it can be accurately identified based on the PDWs at multiple moments, thereby improving the recognition accuracy of the radar working mode; based on the PDW sequence, the intra-pulse characteristics of the radar radiation source and the inter-pulse characteristics of the radar radiation source are determined, so that the radar working mode is accurately identified based on the intra-pulse characteristics and the inter-pulse characteristics, thereby improving the recognition accuracy of the radar working mode; the intra-pulse characteristics and the inter-pulse characteristics are input into the working mode feature extraction model to obtain the intra-pulse feature vector and the inter-pulse feature vector output by the working mode feature extraction model, based on which, further feature extraction is performed on the intra-pulse characteristics and the inter-pulse characteristics respectively, and working mode feature extraction can be performed on the parameters inside the radar pulse and between the radar pulses respectively, thereby ensuring the extraction of more comprehensive working mode features, thereby improving the recognition accuracy of the radar working mode; and the working mode recognition method can be accurately identified based on the intra-pulse characteristics and the inter-pulse characteristics. The working mode feature extraction model is used to extract the intra-pulse feature vector with context information based on the intra-pulse feature, and the working mode feature extraction model is also used to extract the inter-pulse feature vector with context information based on the inter-pulse feature, so as to fully extract the implicit features and time series correlation information within the PDW sequence, thereby improving the recognition accuracy of the radar working mode; based on the input features and feature extraction method determined above, the feature fusion vector of the intra-pulse feature vector and the inter-pulse feature vector is input into the radar working mode recognition model, and an accurate radar working mode recognition result can be obtained. At the same time, the intra-pulse feature and the inter-pulse feature are first obtained based on the PDW sequence, and then the working mode feature extraction is performed on the intra-pulse feature and the inter-pulse feature to obtain the intra-pulse feature vector and the inter-pulse feature vector with context information, so that the radar working mode can be recognized with high precision for complex electromagnetic environment scenarios such as incomplete radar signal parameters or overlapping radar signal parameters, and high precision recognition can be achieved regardless of the scenario. In summary, the present invention can achieve high-precision recognition of the radar working mode for each scenario, that is, the present invention can improve the recognition accuracy of the radar working mode and enhance the adaptability of the radar working mode.

[0058] Based on any of the above embodiments, a specific embodiment of a radar operating mode recognition method is given below. Figure 2 This is the second flow chart of the radar working mode recognition method provided by the present invention, such as Figure 2 As shown, the radar working mode recognition method includes: step 110, step 121, step 122, step 123, step 130 and step 140.

[0059] Step 110: Acquire a pulse descriptor word (PDW) sequence of a radar emitter.

[0060] Here, the PDW sequence includes PDWs at multiple time instants, and the PDW at any time instant includes carrier frequency, pulse width, and arrival time.

[0061] Step 121 : extract the carrier frequency, pulse width and arrival time of the PDW sequence respectively to obtain a carrier frequency sequence, a pulse width sequence and an arrival time sequence.

[0062] Here, the carrier frequency sequence includes carrier frequencies at multiple moments, the pulse width sequence includes pulse widths at multiple moments, and the arrival time sequence includes arrival times at multiple moments.

[0063] Step 122: Concatenate the carrier frequency sequence and the pulse width sequence to obtain the intra-pulse feature.

[0064] Specifically, the carrier frequency sequence and pulse width sequence are concatenated to form an intra-pulse feature. This concatenation of the carrier frequency sequence and pulse width sequence yields an intra-pulse feature that characterizes the signal characteristics and parameters within a single radar pulse. In one embodiment, this concatenated intra-pulse feature is a single feature, which, when used in the operating mode feature extraction model, can still extract an intra-pulse feature vector containing contextual information.

[0065] It should be understood that by splicing the carrier frequency sequence and the pulse width sequence, more accurate intra-pulse features can be obtained, thereby improving the recognition accuracy of the radar operating mode.

[0066] Step 123: performing differential processing on the arrival time series to obtain the inter-pulse features.

[0067] Here, differencing processing can include, but is not limited to, first-order and second-order differencing. For example, a second-order differencing of the arrival time series can be performed to obtain inter-pulse features. Alternatively, a first-order differencing of the arrival time series can be performed, followed by a second-order differencing of the first-order differencing sequence to obtain inter-pulse features. Differencing the arrival time series can yield inter-pulse features that characterize parameter variations between different radar pulses.

[0068] It should be understood that performing differential processing on the arrival time series can obtain more accurate inter-pulse features, thereby improving the recognition accuracy of the radar operating mode.

[0069] Step 130: Input the intra-pulse feature and the inter-pulse feature into a working mode feature extraction model to obtain an intra-pulse feature vector and an inter-pulse feature vector output by the working mode feature extraction model.

[0070] It should be noted that the intra-pulse features and inter-pulse features are input into the working mode feature extraction model to further extract potential semantic features, that is, to extract inter-pulse feature vectors and intra-pulse feature vectors with contextual information, so as to fully extract (mine) the implicit features and temporal correlation information within the PDW sequence, that is, to extract deeper grammatical rules and temporal correlation relationships, that is, to extract more comprehensive and complete features, thereby improving the accuracy of radar working mode recognition.

[0071] In a specific embodiment, the working mode feature extraction model includes two branches to respectively extract the working mode features of intra-pulse features and inter-pulse features.

[0072] Step 140: Input the feature fusion vector of the intra-pulse feature vector and the inter-pulse feature vector into a radar working mode recognition model to obtain a radar working mode recognition result output by the radar working mode recognition model.

[0073] The radar operating mode recognition method provided by the embodiment of the present invention can obtain more accurate intra-pulse features by splicing the carrier frequency sequence and the pulse width sequence, thereby further improving the recognition accuracy of the radar operating mode; and can obtain more accurate inter-pulse features by performing differential processing on the arrival time series, thereby further improving the recognition accuracy of the radar operating mode.

[0074] Based on any of the above embodiments, a specific embodiment of a radar operating mode recognition method is given below. Figure 3 This is the third flow chart of the radar working mode recognition method provided by the present invention, such as Figure 3 As shown, the radar working mode recognition method includes: step 110, step 120, step 131, step 132 and step 140.

[0075] Step 110: Acquire a pulse descriptor word (PDW) sequence of a radar emitter.

[0076] Here, the PDW sequence includes PDWs at multiple moments, and the PDW sequence is a sequence to be identified in the radar working mode.

[0077] Step 120: Determine the intra-pulse characteristics and the inter-pulse characteristics of the radar emitter based on the PDW sequence.

[0078] It should be noted that the PDW sequence includes timing information, so the intra-pulse features and inter-pulse features still imply the timing information of the PDW sequence. For the working mode feature extraction model, it can still extract intra-pulse feature vectors and inter-pulse feature vectors with context information.

[0079] Step 131: Input the intra-pulse feature into the first long short-term memory network layer in the working mode feature extraction model to obtain the intra-pulse feature vector output by the first long short-term memory network layer.

[0080] Here, the first long short-term memory network layer utilizes the superior expressive power of the deep learning network to extract deep rules, that is, to extract the implicit features within the PDW sequence, thereby improving the accuracy of radar working mode recognition; and the first long short-term memory network layer is used to extract intra-pulse feature vectors with contextual information based on intra-pulse features, that is, it is used to further extract working mode features with contextual information, thereby fully extracting (mining) the implicit features and temporal correlation information within the PDW sequence, that is, extracting deeper grammatical rules and temporal correlation relationships, that is, extracting more comprehensive and complete features, thereby improving the accuracy of radar working mode recognition.

[0081] Furthermore, the intra-pulse feature is input into the encoding layer of the working mode feature extraction model to obtain a first encoding vector output by the encoding layer. The first encoding vector is then input into the first long short-term memory network layer of the working mode feature extraction model to obtain an intra-pulse feature vector output by the first long short-term memory network layer. The encoding layer can be configured as needed, for example, including five one-dimensional convolutional layers.

[0082] Step 132: Input the inter-pulse feature into the second long short-term memory network layer in the working mode feature extraction model to obtain the inter-pulse feature vector output by the second long short-term memory network layer.

[0083] Here, the second long short-term memory network layer utilizes the superior expressive power of the deep learning network to extract deep rules, that is, to extract the implicit features within the PDW sequence, thereby improving the accuracy of radar working mode recognition; and the second long short-term memory network layer is used to extract inter-pulse feature vectors with contextual information based on inter-pulse features, that is, it is used to further extract working mode features with contextual information, thereby fully extracting (mining) the implicit features and temporal correlation information within the PDW sequence, that is, extracting deeper grammatical rules and temporal correlation relationships, that is, extracting more comprehensive and complete features, thereby improving the accuracy of radar working mode recognition.

[0084] Furthermore, the inter-pulse feature is input into the encoding layer of the working mode feature extraction model to obtain a second encoding vector output by the encoding layer. The second encoding vector is then input into the second long short-term memory network layer of the working mode feature extraction model to obtain an inter-pulse feature vector output by the second long short-term memory network layer. The encoding layer can be configured as needed; for example, the encoding layer may include five one-dimensional convolutional layers.

[0085] The working mode feature extraction model is obtained by unsupervised training based on a sample PDW sequence. The sample PDW sequence is a training sample, and the sample PDW sequence can refer to the above-mentioned PDW sequence. Unsupervised training methods may include, but are not limited to, contrastive learning, generative adversarial learning, deep clustering, and the like.

[0086] It should be understood that the working mode feature extraction model is obtained through unsupervised training, so there is no need for labeling, which can reduce the labeling cost; and it can make full use of the massive unlabeled sample PDW sequence for model training, thereby effectively improving the performance of the model and ultimately improving the accuracy of radar working mode recognition.

[0087] Step 140: Input the feature fusion vector of the intra-pulse feature vector and the inter-pulse feature vector into a radar working mode recognition model to obtain a radar working mode recognition result output by the radar working mode recognition model.

[0088] The radar working mode recognition method provided by the embodiment of the present invention further extracts intra-pulse features and inter-pulse features through the first long short-term memory network layer and the second long short-term memory network layer, respectively, and can extract context information of features within the radar pulse and between radar pulses respectively, thereby ensuring the extraction of more comprehensive working mode features, thereby improving the accuracy of radar working mode recognition; and the working mode feature extraction model is obtained by unsupervised training based on the sample PDW sequence, thereby reducing the labeling cost and making full use of massive unlabeled sample PDW sequences for model training, thereby improving the robustness of the working mode feature extraction model, and ultimately improving the recognition accuracy of the radar working mode.

[0089] Based on any of the above embodiments, a specific embodiment of a radar working mode recognition method is given below: The working mode feature extraction model is a two-way contrast prediction coding model, and the working mode feature extraction model is used to extract essential features.

[0090] Here, a bidirectional contrastive predictive coding model is constructed based on the BiCPC (Bidirectional Contrastive Predictive Coding) method to extract the potential semantic features of the PDW sequence.

[0091] The BiCPC method is used to construct the working mode feature extraction model because it is a self-supervised learning method that combines bidirectional context modeling and contrastive predictive coding (CPC). It aims to learn richer sequence data representations by simultaneously utilizing past and future contextual information. It is also a high-level representation extraction technology based on unsupervised learning, and has achieved excellent performance in multiple fields such as text, images, and speech, thereby improving the accuracy of radar working mode recognition.

[0092] Exemplarily, the working mode feature extraction model is trained based on the following method: The first step is to determine the intra-pulse features and inter-pulse features of the samples at time t based on the sample PDW sequence at time t; The second step is to input the intra-pulse features and inter-pulse features of the sample at time t into the coding layer of the working mode feature extraction model, and obtain the first coding vector and the second coding vector of the sample at time t output by the coding layer; The third step is to input the first encoding vector and the second encoding vector of the sample at time t into the LSTM layer of the working mode feature extraction model, and obtain the intra-pulse feature vector and inter-pulse feature vector of the sample at time t output by the LSTM layer; wherein the LSTM layer determines the intra-pulse feature vector and inter-pulse feature vector of the sample at time t based on the encoding vector at time t and the state output vector at the previous time t-1; The fourth step is to predict the coding vectors for the next n moments at time t based on the intra-pulse feature vector and inter-pulse feature vector of the sample at time t, that is, to obtain the first coding vector and the second coding vector of the n samples from time t+1 to time t+n. The prediction can be performed through a linear layer. Step 5: Calculate the similarity between the predicted sample first coding vector and the sample second coding vector and the real sample first coding vector and the sample second coding vector; In the sixth step, the loss value is determined based on the similarity calculation results, and the working mode feature extraction model is trained based on the loss value.

[0093] In other words, let the set of sample PDW sequences of a single training batch be {X(i)}, the number of samples in the sample PDW sequence set be M, and the contrastive learning training goal (training criterion) is to predict the similarity scores of the coding vectors of the sample PDW sequence X(i) in the future n moments through the sample intra-pulse feature vectors and the sample inter-pulse feature vectors of the sample PDW sequence X(i) to be higher and higher, while the similarity scores of the coding vectors of other sample PDW sequences in the future n moments to be lower and lower.

[0094] For example, the loss function of the working mode feature extraction model is as follows: ; Where, represents the number of future moments (the number of future time steps to predict), is a positive integer greater than 0; Represents a similarity calculation function (such as a cosine similarity calculation function); Represents the future time of the sample PDW sequence X(i) The predictive coding vector of Represents the future time of the sample PDW sequence X(i) The true encoding vector of Represents a set of sample PDW sequences for a single training batch.

[0095] It should be noted that the bidirectional contrastive predictive coding model uses the code vectors at the current and past moments to predict the code vectors at future moments. This effectively extracts an essential feature (background feature) in the PDW sequence that is stable throughout the entire signal phase, thereby improving the accuracy of radar operating mode recognition. Furthermore, through the aforementioned contrastive learning training principle, where the current sample can only predict the code vector for the current sample at future moments, but not for other samples in the same training batch, it is able to extract discriminative information about the PDW sequence. Combined with the ability to extract the essential features (background features) of the PDW sequence data, the two methods are combined to extract a highly discriminative operating mode feature of the PDW sequence, thereby improving the accuracy of radar operating mode recognition.

[0096] In the radar operating mode recognition method provided by an embodiment of the present invention, the operating mode feature extraction model is a two-way contrast prediction coding model, thereby using the coding vectors at the current moment and the past moment to predict the coding vector at the future moment. This can effectively extract an essential feature in the PDW sequence that is stable throughout the entire signal stage, thereby improving the feature extraction capability of the operating mode feature extraction model, thereby improving the recognition accuracy of the radar operating mode. The two-way contrast prediction coding model can extract a highly distinguishable operating mode feature of the PDW sequence, thereby improving the recognition accuracy of the radar operating mode.

[0097] Based on any of the above embodiments, another embodiment of the radar operating mode recognition method is given below. Figure 4 FIG4 is a flow chart of the radar working mode recognition method provided by the present invention, as shown in FIG4. Figure 4 As shown, the radar working mode recognition method includes: step 110, step 120, step 130, step 140, step 150 and step 160. Step 150 is performed after step 110, and step 160 is performed after step 140.

[0098] Step 150: Input the PDW sequence into a radar operating mode detection model to obtain a radar operating mode detection result output by the radar operating mode detection model.

[0099] The radar operating mode detection model is used to perform radar operating mode detection based on the PDW sequence.

[0100] In one embodiment, the radar operating mode detection model includes multiple machine learning classifiers, eliminating the need for template matching and improving the accuracy of radar operating mode recognition. The radar operating mode detection model can be constructed based on algorithms such as support vector machines (SVMs), random forests, and decision trees.

[0101] It should be noted that the traditional machine learning pattern recognition method is based on artificially defined physical features and has good interpretability. Compared with the deep learning method (that is, the method of obtaining the radar working mode recognition results mentioned above), machine learning can also have a certain effect on working modes with a small number of samples and is not completely unusable. It is very suitable as a backup model for radar working mode recognition. Therefore, the radar working mode detection model is also used for radar working mode detection.

[0102] In one specific embodiment, the radar operating mode detection model is built based on a random forest algorithm. The radar operating mode detection model is trained by randomly extracting multiple training sets from a prior sample set, then building multiple decision tree classifiers using the training sets. Finally, the multiple decision tree classifiers identify and vote on feature vectors (i.e., parameters in the PDW sequence) to produce an operating mode detection result.

[0103] In one embodiment, a carrier frequency sequence, a pulse width sequence, and a pulse repetition period sequence are determined based on a PDW sequence. These sequences are then input into a radar operating mode detection model to obtain a radar operating mode detection result output by the radar operating mode detection model. It should be understood that using the carrier frequency sequence, pulse width sequence, and pulse repetition period sequence as identification feature vectors for the radar operating mode detection model can improve the accuracy of radar operating mode detection.

[0104] Here, the radar working mode detection result is used to indicate the radar working mode of the radar radiation source. The radar working mode may include but is not limited to: speed search, tracking plus search, tracking while ranging, multi-target tracking, tracking while searching, and single target tracking, etc.

[0105] Step 160: Determine a final radar operating mode recognition result based on the radar operating mode recognition result and the radar operating mode detection result.

[0106] Specifically, based on the radar working mode recognition result and the radar working mode detection result, a comprehensive judgment is made to obtain the final recognition result of the radar working mode.

[0107] Exemplarily, assuming that the total number of radar operating modes is M, the first confidences of the M radar operating modes are determined based on the radar operating mode recognition results, the second confidences of the M radar operating modes are determined based on the radar operating mode detection results, and the comprehensive confidences of the M radar operating modes are determined based on the M first confidences and the M second confidences (the comprehensive confidence of any radar operating mode is obtained by fusing the first confidence and the second confidence of the radar operating mode, and the fusion method can be averaging or weighted averaging, etc.). Based on the comprehensive confidences of the M radar operating modes, the final recognition result of the radar operating mode is determined, that is, the radar operating mode with the highest comprehensive confidence is determined as the final recognition result of the radar operating mode.

[0108] Here, the final identification result of the radar working mode is used to indicate the radar working mode of the radar radiation source. The radar working mode may include but is not limited to: speed search, tracking plus search, tracking while ranging, multi-target tracking, tracking while searching, and single target tracking, etc.

[0109] It should be noted that, through experimental verification, the radar working mode recognition effect is significantly improved by using both the radar working mode recognition model and the radar working mode detection model compared to using only the radar working mode recognition model.

[0110] The radar operating mode recognition method provided by an embodiment of the present invention also inputs the PDW sequence into a radar operating mode detection model to obtain a radar operating mode detection result output by the radar operating mode detection model, so as to comprehensively determine the final recognition result of the radar operating mode based on the radar operating mode recognition result and the radar operating mode detection result, thereby adopting two methods to identify the radar operating mode, thereby improving the recognition accuracy of the radar operating mode.

[0111] Based on any of the above embodiments, another embodiment of the radar operating mode recognition method is given below. Figure 5 FIG5 is a flow chart of the radar working mode recognition method provided by the present invention, as shown in FIG5. Figure 5 As shown, the radar working mode recognition method includes: step 110, step 120, step 130, step 140, step 151 and step 160. Step 151 is after step 110, and step 160 is after step 140.

[0112] Step 151: Input each PDW in the PDW sequence into a radar operating mode detection model to obtain a radar operating mode detection result output by the radar operating mode detection model.

[0113] The radar operating mode detection model is used to perform radar operating mode detection based on a PDW.

[0114] In one embodiment, any PDW includes carrier frequency, pulse width, and pulse repetition period. Based on this, the carrier frequency, pulse width, and pulse repetition period are input into a radar operating mode detection model to obtain a radar operating mode detection sub-result output by the radar operating mode detection model. It should be understood that using carrier frequency, pulse width, and pulse repetition period as identification feature vectors of the radar operating mode detection model can improve the accuracy of radar operating mode detection. Of course, other parameters can also be input into the radar operating mode detection model, for example, carrier frequency, pulse width, time of arrival, and direction of arrival can be input into the radar operating mode detection model to obtain a radar operating mode detection sub-result output by the radar operating mode detection model.

[0115] The radar working mode detection result includes the radar working mode detection sub-result corresponding to each PDW, that is, a PDW is input into the radar working mode detection model to obtain a radar working mode detection sub-result output by the radar working mode detection model.

[0116] Here, any radar operating mode detection sub-result can be used to indicate the radar operating mode of the radar radiation source, and the radar operating mode may include but is not limited to: speed search, tracking plus search, tracking while ranging, multi-target tracking, tracking while searching, and single target tracking, etc.

[0117] It should be noted that only a single PDW needs to be input into the radar working mode detection model. The radar working mode detection model still has strong discrimination for working modes with large differences in radar signal parameters, and it does not need to extract timing correlation information.

[0118] Step 160: Determine a final radar operating mode recognition result based on the radar operating mode recognition result and the radar operating mode detection result.

[0119] Specifically, based on the radar working mode recognition result and each radar working mode detection sub-result, a comprehensive judgment is made to obtain the final radar working mode recognition result.

[0120] In a specific embodiment, based on the detection sub-results of each radar working mode, the working mode detection result is comprehensively determined, and based on the radar working mode recognition result and the working mode detection result, a comprehensive judgment is made to obtain the final recognition result of the radar working mode.

[0121] Exemplarily, assuming that the total number of radar operating modes is M and the number of radar operating mode detection sub-results is N, based on the detection sub-results of each radar operating mode, N*M confidences are determined, based on the N*M confidences, the second confidences of the M radar operating modes are determined, based on the radar operating mode recognition results, the first confidences of the M radar operating modes are determined, based on the M first confidences and the M second confidences, the comprehensive confidences of the M radar operating modes are determined (the comprehensive confidence of any radar operating mode is obtained by fusing the first confidence and the second confidence of the radar operating mode, and the fusion method can be averaging or weighted averaging, etc.), based on the comprehensive confidences of the M radar operating modes, the final recognition result of the radar operating mode is determined, that is, the radar operating mode with the highest comprehensive confidence is determined as the final recognition result of the radar operating mode.

[0122] The radar operating mode recognition method provided by the embodiment of the present invention only requires inputting a single PDW into the radar operating mode detection model. The radar operating mode detection model still has strong discrimination for operating modes with large differences in radar signal parameters. Therefore, the radar operating mode detection model can serve as a backup model for radar operating mode recognition, thereby improving the recognition accuracy of radar operating modes.

[0123] To facilitate understanding of the above embodiments, a specific embodiment is used for illustration. Figure 6 As shown, in the first branch, based on the PDW sequence, the carrier frequency sequence, pulse width sequence and arrival time sequence are extracted, based on the carrier frequency sequence and the pulse width sequence, the intra-pulse feature is determined, based on the arrival time sequence, the inter-pulse feature is determined, then the intra-pulse feature is input into the coding layer and the first long short-term memory network layer in the working mode feature extraction model to obtain the intra-pulse feature vector output by the first long short-term memory network layer, and the inter-pulse feature is input into the coding layer and the second long short-term memory network layer in the working mode feature extraction model to obtain the inter-pulse feature vector output by the second long short-term memory network layer, then the feature fusion vector of the intra-pulse feature vector and the inter-pulse feature vector is input into the radar working mode recognition model to obtain the radar working mode recognition result output by the radar working mode recognition model; in the second branch, each PDW in the PDW sequence is input into the radar working mode detection model respectively to obtain each radar working mode detection sub-result output by the radar working mode detection model; finally, based on the radar working mode recognition result and each radar working mode detection sub-result, the final recognition result of the radar working mode is determined.

[0124] The intra-pulse feature vector and the inter-pulse feature vector are input into the working mode feature extraction model to obtain the intra-pulse feature vector and the inter-pulse feature vector output by the working mode feature extraction model. The present invention addresses the complex scenarios that may occur in electromagnetic environments, namely, the problem that the multi-function radar has a significant impact on the pulse signal in a certain operating mode, thereby exacerbating the parameter overlap problem, and the problem of low radar operating mode recognition accuracy due to a low signal-to-noise ratio. By adopting the above-mentioned embodiments, the recognition accuracy and robustness of the radar operating mode in complex electromagnetic environments are effectively improved.

[0125] The radar operating mode recognition device provided by the present invention is described below. The radar operating mode recognition device described below and the radar operating mode recognition method described above can be referenced to each other.

[0126] Figure 7 FIG. 1 is a schematic diagram of the structure of the radar working mode recognition device provided by the present invention. Figure 7 As shown, the radar working mode recognition device includes: a sequence acquisition module 710, a feature determination module 720, a feature extraction module 730 and a pattern recognition module 740.

[0127] The sequence acquisition module 710 is used to acquire a pulse descriptor word PDW sequence of a radar emitter; the PDW sequence includes PDWs at multiple moments.

[0128] The feature determination module 720 is configured to determine the intra-pulse feature and the inter-pulse feature of the radar emitter based on the PDW sequence.

[0129] The feature extraction module 730 is used to input the intra-pulse features and the inter-pulse features into the working mode feature extraction model to obtain the intra-pulse feature vector and the inter-pulse feature vector output by the working mode feature extraction model; the working mode feature extraction model is used to extract the intra-pulse feature vector with context information based on the intra-pulse features, and the working mode feature extraction model is also used to extract the inter-pulse feature vector with context information based on the inter-pulse features.

[0130] The pattern recognition module 740 is used to input the feature fusion vector of the intra-pulse feature vector and the inter-pulse feature vector into the radar working mode recognition model to obtain the radar working mode recognition result output by the radar working mode recognition model; the radar working mode recognition model is used to perform radar working mode recognition on the feature fusion vector.

[0131] The radar working mode recognition device provided by the embodiment of the present invention obtains the PDW sequence of the radar radiation source, and the PDW sequence includes PDWs at multiple moments, so that even if the PDW at some moments includes erroneous parameters, it can be accurately identified based on the PDWs at multiple moments, thereby improving the recognition accuracy of the radar working mode; based on the PDW sequence, the intra-pulse characteristics of the radar radiation source and the inter-pulse characteristics of the radar radiation source are determined, so that the radar working mode is accurately identified based on the intra-pulse characteristics and the inter-pulse characteristics, thereby improving the recognition accuracy of the radar working mode; the intra-pulse characteristics and the inter-pulse characteristics are input into the working mode feature extraction model to obtain the intra-pulse feature vector and the inter-pulse feature vector output by the working mode feature extraction model, based on which, further feature extraction is performed on the intra-pulse characteristics and the inter-pulse characteristics respectively, and working mode feature extraction can be performed on the parameters inside the radar pulse and between the radar pulses respectively, thereby ensuring the extraction of more comprehensive working mode features, thereby improving the recognition accuracy of the radar working mode; and the working mode recognition device can be accurately identified based on the intra-pulse characteristics and the inter-pulse characteristics. The working mode feature extraction model is used to extract the intra-pulse feature vector with context information based on the intra-pulse feature, and the working mode feature extraction model is also used to extract the inter-pulse feature vector with context information based on the inter-pulse feature, so as to fully extract the implicit features and time series correlation information within the PDW sequence, thereby improving the recognition accuracy of the radar working mode; based on the input features and feature extraction method determined above, the feature fusion vector of the intra-pulse feature vector and the inter-pulse feature vector is input into the radar working mode recognition model, and an accurate radar working mode recognition result can be obtained. At the same time, the intra-pulse feature and the inter-pulse feature are first obtained based on the PDW sequence, and then the working mode feature extraction is performed on the intra-pulse feature and the inter-pulse feature to obtain the intra-pulse feature vector and the inter-pulse feature vector with context information, so that the radar working mode can be recognized with high precision for complex electromagnetic environment scenarios such as incomplete radar signal parameters or overlapping radar signal parameters, and high precision recognition can be achieved regardless of the scenario. In summary, the present invention can achieve high-precision recognition of the radar working mode for each scenario, that is, the present invention can improve the recognition accuracy of the radar working mode and enhance the adaptability of the radar working mode.

[0132] Based on any of the above embodiments, any of the PDWs includes carrier frequency, pulse width, and arrival time; the feature determination module 720 is specifically configured to: Extracting the carrier frequency, pulse width and arrival time of the PDW sequence respectively to obtain a carrier frequency sequence, a pulse width sequence and an arrival time sequence; splicing the carrier frequency sequence and the pulse width sequence to obtain the intra-pulse feature; The arrival time series is subjected to differential processing to obtain the inter-pulse features.

[0133] Based on any of the above embodiments, the feature extraction module 730 is specifically configured to: Inputting the intra-pulse feature into the first long short-term memory network layer in the working mode feature extraction model to obtain the intra-pulse feature vector output by the first long short-term memory network layer; Inputting the inter-pulse feature into a second long short-term memory network layer in the working mode feature extraction model to obtain an inter-pulse feature vector output by the second long short-term memory network layer; The working mode feature extraction model is obtained by performing unsupervised training based on the sample PDW sequence.

[0134] Based on any of the above embodiments, the working mode feature extraction model is a two-way contrast prediction coding model, and the working mode feature extraction model is used to extract essential features.

[0135] Based on any of the above embodiments, the device further includes: a pattern detection module and a result determination module.

[0136] A mode detection module is used to input the PDW sequence into a radar working mode detection model to obtain a radar working mode detection result output by the radar working mode detection model; the radar working mode detection model is used to perform radar working mode detection based on the PDW sequence.

[0137] A result determination module is used to determine a final recognition result of the radar working mode based on the radar working mode recognition result and the radar working mode detection result.

[0138] Based on any of the above embodiments, the pattern detection module is specifically used to: Each PDW in the PDW sequence is input into a radar working mode detection model to obtain a radar working mode detection result output by the radar working mode detection model.

[0139] The radar operating mode detection model is used to perform radar operating mode detection based on one PDW, and the radar operating mode detection result includes radar operating mode detection sub-results corresponding to each PDW.

[0140] Figure 8 An example of a physical structure diagram of an electronic device is shown below. Figure 8As shown, the electronic device may include: a processor 810 , a communication interface 820 , a memory 830 and a communication bus 840 , wherein the processor 810 , the communication interface 820 and the memory 830 communicate with each other via the communication bus 840 . The processor 810 can call the logic instructions in the memory 830 to execute the radar working mode recognition method, which includes: obtaining a pulse descriptor word PDW sequence of the radar radiation source; the PDW sequence includes PDWs at multiple moments; based on the PDW sequence, determining the intra-pulse characteristics of the radar radiation source and the inter-pulse characteristics of the radar radiation source; inputting the intra-pulse characteristics and the inter-pulse characteristics into the working mode feature extraction model to obtain the intra-pulse feature vector and the inter-pulse feature vector output by the working mode feature extraction model; the working mode feature extraction model is used to extract the intra-pulse feature vector with context information based on the intra-pulse characteristics, and the working mode feature extraction model is also used to extract the inter-pulse feature vector with context information based on the inter-pulse characteristics; inputting the feature fusion vector of the intra-pulse feature vector and the inter-pulse feature vector into the radar working mode recognition model to obtain the radar working mode recognition result output by the radar working mode recognition model; the radar working mode recognition model is used to perform radar working mode recognition on the feature fusion vector.

[0141] Furthermore, the logic instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product, stored in a storage medium, includes instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0142] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the radar working mode recognition method provided by the above methods, the method including: obtaining a pulse descriptor word PDW sequence of a radar radiation source; the PDW sequence includes PDWs at multiple moments; based on the PDW sequence, determining the intra-pulse characteristics and the inter-pulse characteristics of the radar radiation source; inputting the intra-pulse characteristics and the inter-pulse characteristics into a working mode feature extraction model to obtain an intra-pulse feature vector and an inter-pulse feature vector output by the working mode feature extraction model; the working mode feature extraction model is used to extract an intra-pulse feature vector with context information based on the intra-pulse characteristics, and the working mode feature extraction model is also used to extract an inter-pulse feature vector with context information based on the inter-pulse characteristics; inputting a feature fusion vector of the intra-pulse feature vector and the inter-pulse feature vector into a radar working mode recognition model to obtain a radar working mode recognition result output by the radar working mode recognition model; the radar working mode recognition model is used to perform radar working mode recognition on the feature fusion vector.

[0143] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the radar working mode recognition method provided by the above-mentioned methods, the method comprising: obtaining a pulse descriptor word PDW sequence of a radar radiation source; the PDW sequence comprises PDWs at multiple moments; based on the PDW sequence, determining the intra-pulse characteristics and the inter-pulse characteristics of the radar radiation source; inputting the intra-pulse characteristics and the inter-pulse characteristics into a working mode feature extraction model to obtain an intra-pulse feature vector and an inter-pulse feature vector output by the working mode feature extraction model; the working mode feature extraction model is used to extract an intra-pulse feature vector with context information based on the intra-pulse characteristics, and the working mode feature extraction model is also used to extract an inter-pulse feature vector with context information based on the inter-pulse characteristics; inputting a feature fusion vector of the intra-pulse feature vector and the inter-pulse feature vector into a radar working mode recognition model to obtain a radar working mode recognition result output by the radar working mode recognition model; the radar working mode recognition model is used to perform radar working mode recognition on the feature fusion vector.

[0144] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0145] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0146] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A radar working mode recognition method, characterized in that: include: Acquire a pulse descriptor word PDW sequence of a radar radiation source; the PDW sequence includes PDWs at multiple moments; determining, based on the PDW sequence, intra-pulse characteristics of the radar emitter and inter-pulse characteristics of the radar emitter; Inputting the intra-pulse feature and the inter-pulse feature into a working mode feature extraction model to obtain an intra-pulse feature vector and an inter-pulse feature vector output by the working mode feature extraction model; the working mode feature extraction model is used to extract an intra-pulse feature vector with context information based on the intra-pulse feature, and the working mode feature extraction model is further used to extract an inter-pulse feature vector with context information based on the inter-pulse feature; The feature fusion vector of the intra-pulse feature vector and the inter-pulse feature vector is input into a radar working mode recognition model to obtain a radar working mode recognition result output by the radar working mode recognition model; the radar working mode recognition model is used to perform radar working mode recognition on the feature fusion vector.

2. The radar operating mode recognition method according to claim 1, characterized in that: Any of the PDWs includes carrier frequency, pulse width, and arrival time; The determining, based on the PDW sequence, the intra-pulse characteristics and the inter-pulse characteristics of the radar emitter includes: Extracting the carrier frequency, pulse width and arrival time of the PDW sequence respectively to obtain a carrier frequency sequence, a pulse width sequence and an arrival time sequence; splicing the carrier frequency sequence and the pulse width sequence to obtain the intra-pulse feature; The arrival time series is subjected to differential processing to obtain the inter-pulse features.

3. The radar operating mode recognition method according to claim 1, characterized in that: Inputting the intra-pulse feature and the inter-pulse feature into a working mode feature extraction model to obtain an intra-pulse feature vector and an inter-pulse feature vector output by the working mode feature extraction model includes: Inputting the intra-pulse feature into the first long short-term memory network layer in the working mode feature extraction model to obtain the intra-pulse feature vector output by the first long short-term memory network layer; Inputting the inter-pulse feature into a second long short-term memory network layer in the working mode feature extraction model to obtain an inter-pulse feature vector output by the second long short-term memory network layer; The working mode feature extraction model is obtained by performing unsupervised training based on the sample PDW sequence.

4. The radar operating mode recognition method according to claim 3, characterized in that: The working mode feature extraction model is a two-way contrast prediction coding model, and the working mode feature extraction model is used to extract essential features.

5. The radar operating mode recognition method according to any one of claims 1 to 4, characterized in that: After obtaining the pulse descriptor word PDW sequence of the radar radiation source, the method further includes: Inputting the PDW sequence into a radar operating mode detection model to obtain a radar operating mode detection result output by the radar operating mode detection model; the radar operating mode detection model is used to perform radar operating mode detection based on the PDW sequence; After inputting the feature fusion vector of the intra-pulse feature vector and the inter-pulse feature vector into the radar working mode recognition model to obtain the radar working mode recognition result output by the radar working mode recognition model, the method further includes: Based on the radar working mode recognition result and the radar working mode detection result, a final radar working mode recognition result is determined.

6. The radar operating mode recognition method according to claim 5, characterized in that: Inputting the PDW sequence into a radar operating mode detection model to obtain a radar operating mode detection result output by the radar operating mode detection model includes: Inputting each PDW in the PDW sequence into a radar working mode detection model, respectively, to obtain a radar working mode detection result output by the radar working mode detection model; The radar operating mode detection model is used to perform radar operating mode detection based on one PDW, and the radar operating mode detection result includes radar operating mode detection sub-results corresponding to each PDW.

7. A radar working mode recognition device, characterized in that: include: A sequence acquisition module is used to acquire a pulse description word PDW sequence of a radar emitter; the PDW sequence includes PDWs at multiple moments; a feature determination module, configured to determine an intra-pulse feature and an inter-pulse feature of the radar emitter based on the PDW sequence; a feature extraction module, configured to input the intra-pulse feature and the inter-pulse feature into a working mode feature extraction model to obtain an intra-pulse feature vector and an inter-pulse feature vector output by the working mode feature extraction model; the working mode feature extraction model is configured to extract an intra-pulse feature vector having contextual information based on the intra-pulse feature, and the working mode feature extraction model is further configured to extract an inter-pulse feature vector having contextual information based on the inter-pulse feature; The pattern recognition module is used to input the feature fusion vector of the intra-pulse feature vector and the inter-pulse feature vector into the radar working mode recognition model to obtain the radar working mode recognition result output by the radar working mode recognition model; the radar working mode recognition model is used to perform radar working mode recognition on the feature fusion vector.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the radar operating mode recognition method according to any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the radar operating mode recognition method according to any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the radar operating mode recognition method according to any one of claims 1 to 6 is implemented.

Images