Non-ideal observation radar PRI modulation identification method

Abstract

The invention discloses a non-ideal observation radar PRI modulation identification method. The method comprises the following steps: presetting parameter ranges of different PRI modulation types; generating a plurality of PRI pulse sequences for each PRI modulation type; preprocessing a PRI pulse sequence generated by each PRI modulation type, generating a GASF coding image and a GADF coding image, and establishing an identification data set; establishing a PRI modulation recognition model, and fusing the features output by the two convolutional neural networks to obtain a network recognition result; for a PRI pulse sequence corresponding to each sample, determining a continuity identification result of a PRI modulation type by calculating a continuity parameter of the PRI pulse sequence and based on a prior parameter interval; combining the network identification result and the continuity identification result, if the continuity identification result contains the network identification result, taking the network identification result as a final identification result; and acquiring a PRI pulse sequence to be identified, and determining a PRI modulation type by utilizing the trained PRI modulation identification model and combining a calculation result of the continuity parameters of the PRI modulation identification model.

Description

Technical Field

[0001] The present invention relates to the field of radar electronic reconnaissance, and in particular to a PRI modulation recognition method for non-ideal observation radar. Background Art

[0002] The pulse repetition interval (PRI) is the interval between the time of arrival (TOA) of adjacent radar pulses and is the reciprocal of the pulse repetition frequency (PRF). To improve radar performance and counter-reconnaissance capabilities, various radar PRI modulation types have emerged. Because PRI has many learnable fixed patterns and is directly related to radar performance and operating system, identifying the PRI modulation type of a radar emitter facilitates understanding its operating mode and mission. However, the diversity of PRI modulation methods, coupled with interference from factors such as pulse loss, false pulses, and external noise caused by non-ideal observation of radar emitters, makes accurate identification of PRI modulation types still a significant challenge.

[0003] PRI modulation type identification methods fall into two main categories. One type performs statistical processing on the TOA sequence, such as extracting its first-order difference features, and then uses manual experience to determine the PRI modulation type based on the histogram distribution. This method relies on artificial features and cannot identify complex PRI modulation patterns. The other type uses artificial intelligence techniques to extract and classify time-domain TOA features. For example, PRI modulation identification methods based on convolutional neural networks and long short-term memory networks use deep learning methods to exploit the periodic characteristics between PRI sequences. This method has high recognition accuracy when pulse detection is discontinuous and the signal-to-noise ratio is low.

[0004] Because deep learning methods introduce feature transformation, some time-domain information in the original pulse sequence is lost. This prevents the algorithm from fully exploiting the temporal regularity of the PRI pulse sequence, leading to low recognition accuracy in scenarios with high pulse loss and noisy pulses. Therefore, a new method for identifying PRI modulation types is urgently needed that can preserve the regularity of the original PRI feature sequence under non-ideal observation conditions. Summary of the Invention

[0005] The purpose of the present invention is to provide a non-ideal observation radar PRI modulation recognition method to overcome the problem of low recognition accuracy in the prior art.

[0006] In order to achieve the above tasks, the present invention adopts the following technical solutions:

[0007] A non-ideal observation radar PRI modulation identification method, comprising:

[0008] Preset parameter ranges for different PRI modulation types; generate multiple PRI pulse sequences for each PRI modulation type;

[0009] Preprocess the PRI pulse sequence generated by each PRI modulation type, and use the preprocessed PRI pulse sequence to generate GASF coded images and GADF coded images; use the GASF coded images and GADF coded images to establish a recognition data set;

[0010] Constructing a PRI modulation recognition model; the recognition model uses parallel convolutional neural networks, each of which is used to extract features from the GASF encoded image and the GADF encoded image in the sample; the features output by the two convolutional neural networks are fused to obtain a network recognition result;

[0011] For each PRI pulse sequence corresponding to a sample, the continuity parameter is calculated and the continuity recognition result of the PRI modulation type is determined based on the prior parameter interval; the network recognition result and the continuity recognition result are combined. If the continuity recognition result contains the network recognition result, the network recognition result is used as the final recognition result;

[0012] Obtain the PRI pulse sequence to be identified, use the trained PRI modulation identification model, and combine the calculation results of its continuity parameters to determine the PRI modulation type.

[0013] Simulations are performed separately according to the parameter range of each PRI modulation type to obtain multiple PRI pulse sequences.

[0014] Furthermore, the PRI pulse sequence preprocessing process is as follows:

[0015] Normalizing each PRI pulse in the PRI pulse sequence to obtain a normalized PRI pulse sequence;

[0016] The normalized PRI pulse sequence is transformed into a polar coordinate system; wherein each normalized PRI pulse Angle converted to polar coordinates

[0017]

[0018] in, represents the i-th PRI pulse in the normalized PRI pulse sequence, and N is the length of the PRI pulse sequence.

[0019] Furthermore, the preprocessed PRI pulse sequence is used to generate GASF coded images and GADF coded images, which are expressed as:

[0020]

[0021]

[0022] Among them, G GASF , G GADF They are the generated GASF encoded image and GADF encoded image respectively; Represents the nth PRI pulse in the normalized PRI pulse sequence The corresponding angle in polar coordinates.

[0023] Furthermore, for each PRI modulation type, each PRI pulse sequence generates a set of GASF and GADF coded images; a set of GASF and GADF coded images is used as a sample in the dataset, and the label of the sample is the PRI modulation type.

[0024] Furthermore, the parallel convolutional neural networks have the same network structure, all of which are CNN network branches. Each CNN network branch has a 9-layer network structure, with the first, third, fifth to seventh layers being convolutional layers, the second, fourth, and eighth layers being pooling layers, and the ninth layer being a fully connected layer. A ReLU activation function is used after each convolutional layer to introduce nonlinearity.

[0025] The features of GASF-encoded images and GADF-encoded images are extracted respectively through two CNN network branches, and then the flattened features are fused in the feature fusion layer. Finally, the fused features are input into another fully connected layer to realize the classification and recognition of the samples, and the network recognition result of PRI modulation type is obtained.

[0026] Furthermore, the continuity parameter determination process of the PRI pulse sequence is as follows:

[0027] Assume that the continuity threshold of each PRI pulse sequence is:

[0028]

[0029] Among them, x i represents the i-th PRI pulse in the PRI pulse sequence, and the length of the PRI pulse sequence is N;

[0030] The continuity parameter C is defined as the ratio of the number of PRI pulses M that meet the continuity requirement to the total length N of the PRI pulse sequence; the number of PRI pulses M that meet the continuity requirement is x i with x i-1 、x i+1 The number of pulses whose absolute value of the difference is lower than the continuity threshold ε.

[0031] Furthermore, the a priori parameter interval for each PRI modulation type is determined in advance by statistical calculation. Then, each PRI pulse sequence is subjected to continuity parameter calculation. When its continuity parameter falls within the a priori parameter interval of a certain PRI modulation type, the PRI pulse sequence X is considered to be the PRI modulation type corresponding to this interval, and this is used as the continuity recognition result.

[0032] Furthermore, the PRI modulation types include fixed PRI, jitter PRI, staggered PRI, group-variable PRI, sliding PRI, and sinusoidal PRI; wherein:

[0033] The sine PRI parameter range is 100-3000μs, with a staggered number of 5-20, a sliding step of 10-100μs, and a period of 2-20; the group variable PRI parameter range is 100-3000μs, with a staggered number of 3-10, and a period of 10-20; the sliding PRI parameter range is 300-500μs, with a staggered number of 3-20, and a sliding step of 10-100μs; the jitter PRI parameter range is 100-3000μs, with a jitter range of 2-20%; the staggered PRI parameter range is 100-3000μs, with a staggered number of 3-20; the fixed PRI parameter range is 100-3000μs;

[0034] A terminal device includes a processor, a memory, and a computer program stored in the memory; when the processor executes the computer program, the non-ideal observation radar PRI modulation identification method is implemented.

[0035] A computer-readable storage medium stores a computer program; when the computer program is executed by a processor, the non-ideal observation radar PRI modulation identification method is implemented.

[0036] Compared with the prior art, the present invention has the following technical features:

[0037] 1. The one-dimensional PRI sequence is converted from Cartesian coordinates to polar coordinates, and the Gram matrix is ​​encoded based on trigonometric functions to generate a two-dimensional GAF image based on PRI, thereby obtaining temporal correlation and revealing hidden features in the PRI sequence data.

[0038] 2. GASF and GADF are used together to encode the image, and a parallel CNN network is used to extract and fuse feature information, fully extracting the multi-dimensional features of the PRI modulation type;

[0039] 3. This method can still achieve a relatively high recognition accuracy under non-ideal observation conditions with pulse loss and spurious pulse interference. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 Schematic diagram of six PRI modulation types under ideal conditions, where (a) is a fixed PRI pulse sequence, (b) is a group-variable PRI pulse sequence, (c) is a staggered PRI pulse sequence, (d) is a sinusoidal PRI pulse sequence, (e) is a sliding PRI pulse sequence, and (f) is a jittered PRI pulse sequence.

[0041] Figure 2 Images of six PRI modulation types converted into GASF: (a) fixed PRI pulse sequence, (b) group-variable PRI pulse sequence, (c) staggered PRI pulse sequence, (d) sinusoidal PRI pulse sequence, (e) sliding PRI pulse sequence, and (f) jittered PRI pulse sequence.

[0042] Figure 3 Images of six PRI modulation types converted into GADF: (a) fixed PRI pulse sequence, (b) group-variable PRI pulse sequence, (c) staggered PRI pulse sequence, (d) sinusoidal PRI pulse sequence, (e) sliding PRI pulse sequence, and (f) jittered PRI pulse sequence.

[0043] Figure 4 This is a block diagram of the PRI modulation recognition method for non-ideal observation radar based on Gram angle field joint coding;

[0044] Figure 5 It is a parallel CNN network structure diagram;

[0045] Figure 6 is the recognition accuracy under different pulse loss rates and different pulse noise rates; (a) is the recognition accuracy under different loss conditions, (b) is the recognition accuracy under different noise conditions;

[0046] Figure 7 is the confusion matrix under different pulse loss rates; (a) is the confusion matrix under 40% loss rate, (b) is the confusion matrix under 50% loss rate;

[0047] Figure 8 is the confusion matrix under different impulse noise rates; (a) is the confusion matrix under 40% loss rate, and (b) is the confusion matrix under 50% loss rate. DETAILED DESCRIPTION

[0048] The present invention addresses the problem of identifying PRI modulation types of radar emitters and provides a non-ideal observation radar PRI modulation identification method. By transforming a one-dimensional PRI sequence into polar coordinates and utilizing a joint coding method of trigonometric function differences and trigonometric function sums, the temporal correlation characteristics of different PRI modulation types are extracted. This method can achieve high-accuracy identification of six PRI modulation types under non-ideal observation conditions of pulse loss and pulse interference. Referring to the accompanying drawings, the present invention is described in detail as follows:

[0049] Step 1: Preset parameter ranges for different PRI modulation types; generate multiple PRI pulse sequences for each PRI modulation type.

[0050] In radar applications, the common PRI modulation types are fixed PRI, jitter PRI, staggered PRI, group variable PRI, sliding PRI, and sinusoidal PRI. The parameter ranges of the PRI modulation types of the present invention are set as follows:

[0051] The sine PRI parameter range is 100-3000μs, with a staggered number of 5-20, a sliding step of 10-100μs, and a period number of 2-20; the group variable PRI parameter range is 100-3000μs, with a staggered number of 3-10, and a period number of 10-20; the sliding PRI parameter range is 300-500μs, with a staggered number of 3-20, and a sliding step of 10-100μs; the jitter PRI parameter range is 100-3000μs, with a jitter range of 2-20%; the staggered PRI parameter range is 100-3000μs, with a staggered number of 3-20; and the fixed PRI parameter range is 100-3000μs.

[0052] Table 1 PRI parameter settings

[0053]

[0054]

[0055] According to the parameter range of each PRI modulation type mentioned above, simulations are performed to obtain multiple PRI pulse sequences of six PRI modulation types: fixed, jitter, staggered, group-variable, sliding, and sine. Among them, a PRI pulse sequence of each modulation type is denoted as X = {x1, x2, ..., x N}, where x i represents the i-th PRI pulse, characterized by TOA; the length of the PRI pulse sequence is N.

[0056] Step 2: Preprocess the PRI pulse sequence generated by each PRI modulation type, and use the preprocessed PRI pulse sequence to generate GASF encoded images and GADF encoded images; use the GASF encoded images and GADF encoded images to establish a recognition data set.

[0057] Gramian Angle Field (GAF) is a method for feature encoding of one-dimensional time series. This method converts the signal in the Cartesian coordinate system to the polar coordinate system for representation, and then obtains information characterizing time correlation in the form of trigonometric function difference and trigonometric function sum according to the inner product definition form, generating two types of Gramian Angular Difference Field (GADF) and Gramian Angular Summation Field (GASF).

[0058] The specific steps for converting PRI pulse sequences into GADF and GASF coded images are as follows:

[0059] Step 21: In order to prevent the inner product of the input sequence from being biased towards the maximum observation value, each PRI pulse sequence X={x1,x2,…,x N} is scaled into the interval [-1,1], the formula is as follows:

[0060]

[0061] in, represents the i-th PRI pulse x i The normalized results, max(X) and min(X), are the maximum and minimum values ​​in the PRI pulse sequence X.

[0062] Step 22, converting the normalized PRI pulse sequence into a polar coordinate system; wherein each normalized PRI pulse Convert to angle in polar coordinates

[0063]

[0064] Step 23: Generate GASF and GADF coded images. The transformation formulas of the two coding methods are as follows:

[0065]

[0066] For each PRI modulation type, each PRI pulse sequence can generate a set of GASF and GADF coded images; a set of GASF and GADF coded images is taken as a sample in the dataset, and the label of the sample is the PRI modulation type.

[0067] The same method is used to generate samples of different PRI modulation types.

[0068] In one embodiment of the present invention, the length of each PRI pulse sequence is N=300, 2200 samples are generated for each PRI modulation type, and the data set is divided into a training set, a test set, and a validation set in a ratio of 7:2:1.

[0069] Step 3, construct a PRI modulation recognition model; the recognition model uses parallel convolutional neural networks, each of which is used to extract features of the GASF encoded image and the GADF encoded image in the sample; the features output by the two convolutional neural networks are fused to obtain the network recognition result.

[0070] Among them, the parallel convolutional neural networks have the same network structure, both of which are CNN network branches, one branch processes GASF encoded images, and the other branch processes GADF encoded images.

[0071] Each CNN network branch has a 9-layer network structure. The first, third, fifth to seventh layers are all convolutional layers, the second, fourth and eighth layers are pooling layers, and the ninth layer is a fully connected layer. Among them, the ReLU activation function is used after each convolutional layer to introduce nonlinearity.

[0072] The first layer is the convolution layer with a convolution kernel size of 7×7, 3 input channels, 48 ​​output channels, and a convolution operation sliding step of 2. It performs primary feature extraction and captures the global features of the encoded image through a large convolution kernel.

[0073] The second layer is the pooling layer, which uses the maximum pooling of a 2×2 window to downsample the input features, thereby reducing the spatial dimension and computational complexity;

[0074] The third layer is the convolution layer, with a convolution kernel size of 5×5, 48 input channels, 128 output channels, and a convolution operation sliding step of 1. It performs intermediate feature extraction and expands the number of channels to enrich feature representation.

[0075] The fourth layer is the pooling layer, which uses the maximum pooling of 2×2 windows and downsamples again to compress the spatial dimension;

[0076] The fifth to seventh layers all use convolutional layers with a convolution kernel size of 3×3. Through advanced feature extraction, small convolution kernels are used to refine features, and continuous convolution is used to compress the number of channels and achieve feature refinement.

[0077] The eighth layer is the pooling layer, which uses the maximum pooling of 2×2 windows to downsample the input features.

[0078] The ninth layer is a fully connected layer that flattens the features output by the pooling layer.

[0079] The features of GASF-encoded images and GADF-encoded images are extracted respectively through two CNN network branches, and then the flattened features are fused in the feature fusion layer. Finally, the fused features are input into another fully connected layer to realize the classification and recognition of the samples, and the network recognition result of PRI modulation type is obtained.

[0080] When training the PRI modulation recognition model, the loss function uses cross-entropy error loss, the minibatch size is 32, the optimization algorithm uses SGD, the learning rate is set to 0.001, and the momentum is set to 0.9; through continuous training and parameter adjustment, until the network performance reaches the best, the current best network model is saved.

[0081] Step 4: For each sample corresponding to the PRI pulse sequence X, the continuity parameter is calculated and the continuity recognition result of the PRI modulation type is determined based on the prior parameter interval;

[0082] The network identification result and the continuity identification result are combined. If the continuity identification result contains the network identification result, the network identification result is used as the final identification result.

[0083] The process of determining the continuity parameter of the PRI pulse sequence X is as follows:

[0084] The i-th PRI pulse x of the PRI pulse sequence X i The condition for the existence of continuity is x i with x i-1 、x i+1 The absolute value of the difference is lower than the continuity threshold ε. Considering the influence of random jitter, the continuity threshold of each PRI pulse sequence X is set to:

[0085]

[0086] The continuity parameter C is defined as the ratio of the number of PRI pulses M that meet the continuity requirement to the total length N of the PRI pulse sequence X, that is:

[0087] C=M / N(6)

[0088] Among them, the number of PRI pulses M that meet the continuity requirement is x i with x i-1 、x i+1 The number of pulses whose absolute value of the difference is lower than the continuity threshold ε is defined as follows:

[0089]

[0090] f(·) is the indicator function, which is defined as follows:

[0091]

[0092] The a priori parameter interval for each PRI modulation type is determined in advance by statistical calculation. Then, each PRI pulse sequence X is subjected to continuity parameter calculation. When its continuity parameter falls within the a priori parameter interval of a certain PRI modulation type, the PRI pulse sequence X is considered to be the PRI modulation type corresponding to this interval and is used as the continuity recognition result.

[0093] Therefore, if the continuity recognition result of a certain PRI pulse sequence includes a network recognition result, the network recognition result is taken as the final recognition result.

[0094] In one embodiment of the present invention, the priori parameter interval of the fixed PRI, sinusoidal PRI and group-variable PRI continuity parameters is 0.8-1, the priori parameter interval of the staggered PRI and sliding PRI continuity parameters is 0-0.05, and the priori parameter interval of the jitter PRI continuity parameter is 0.1-0.3.

[0095] That is, for example, if the network recognition result of a PRI pulse sequence is "fixed PRI", and the calculated value of its continuity parameter C is 0.9, which falls into the prior parameter range of 0.8 to 1, then its continuity recognition result may be one of "fixed PRI, sinusoidal PRI and group-variable PRI", which includes "fixed PRI", so "fixed PRI" is used as the final recognition result.

[0096] Step 5: Obtain the PRI pulse sequence to be identified, cut it into pieces with a length of N, and use the trained PRI modulation identification model to determine its network identification result, and use the continuity parameter to determine its continuity identification result; if the continuity identification result includes the network identification result, then the network identification result is used as the final identification result; otherwise, proceed to identify the next section of the PRI pulse sequence.

[0097] Verify that the method of the present invention has a PRI pulse sequence loss rate R loss =[10%, 20%, 30%, 40%, 50%] and pulse interference R spurious = PRI modulation type recognition accuracy under the conditions of [10%, 20%, 30%, 40%, 50%], the recognition results are as follows Figure 6 As shown in (a) and (b), as the loss rate changes, the recognition accuracy of each category changes, and the overall trend is that the accuracy decreases as the loss rate and noise rate increase.

[0098] The simulation results show that the PRI modulation type is less affected by lost pulses than by noise pulses. When the loss rate reaches 50%, the PRI type recognition accuracy remains above 94%, and when the noise rate reaches 50%, the PRI type recognition accuracy remains above 86%. When the pulse loss rate and noise rate of the radiation source are too large, the number of actual pulses in the PRI sequence is too small to reflect the periodicity of the PRI modulation type, resulting in a decrease in recognition accuracy. Figure 7 and Figure 8 These are the recognition confusion matrices under different loss rates and noise rates.

[0099] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application 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. 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 application, and should all be included in the scope of protection of the present application.

Claims

1. A non-ideal observation radar PRI modulation identification method, characterized in that: include: Preset parameter ranges for different PRI modulation types; Generate multiple PRI pulse trains for each PRI modulation type; Preprocess the PRI pulse sequence generated by each PRI modulation type, and use the preprocessed PRI pulse sequence to generate GASF coded images and GADF coded images; use the GASF coded images and GADF coded images to establish a recognition data set; Constructing a PRI modulation recognition model; the recognition model uses parallel convolutional neural networks, each of which is used to extract features from the GASF encoded image and the GADF encoded image in the sample; the features output by the two convolutional neural networks are fused to obtain a network recognition result; For each PRI pulse sequence corresponding to a sample, the continuity parameter is calculated and the continuity recognition result of the PRI modulation type is determined based on the prior parameter interval. The network recognition result and the continuity recognition result are combined. If the continuity recognition result contains the network recognition result, the network recognition result is used as the final recognition result. Obtain the PRI pulse sequence to be identified, use the trained PRI modulation identification model, and combine the calculation results of its continuity parameters to determine the PRI modulation type. Simulations are performed separately according to the parameter range of each PRI modulation type to obtain multiple PRI pulse sequences.

2. The non-ideal observation radar PRI modulation identification method according to claim 1, characterized in that: The process of preprocessing the PRI pulse sequence is as follows: Normalizing each PRI pulse in the PRI pulse sequence to obtain a normalized PRI pulse sequence; Convert the normalized PRI pulse sequence into the polar coordinate system; Each normalized PRI pulse Angle converted to polar coordinates in, represents the i-th PRI pulse in the normalized PRI pulse sequence, and N is the length of the PRI pulse sequence.

3. The non-ideal observation radar PRI modulation identification method according to claim 1, characterized in that: The GASF coded image and GADF coded image are generated using the preprocessed PRI pulse sequence, which can be expressed as: Among them, G GASF , G GADF They are the generated GASF encoded image and GADF encoded image respectively; Represents the nth PRI pulse in the normalized PRI pulse sequence The corresponding angle in polar coordinates.

4. The non-ideal observation radar PRI modulation identification method according to claim 1, characterized in that: For each PRI modulation type, each PRI pulse sequence generates a set of GASF and GADF coded images; a set of GASF and GADF coded images is used as a sample in the dataset, and the label of the sample is the PRI modulation type.

5. The non-ideal observation radar PRI modulation identification method according to claim 1, characterized in that: The parallel convolutional neural networks have the same network structure, which are all CNN network branches. Each CNN network branch has a 9-layer network structure. The first, third, fifth to seventh layers are all convolutional layers. The second, fourth and eighth layers are pooling layers. The ninth layer is a fully connected layer. The ReLU activation function is used after each convolutional layer to introduce nonlinearity. The features of GASF-encoded images and GADF-encoded images are extracted respectively through two CNN network branches, and then the flattened features are fused in the feature fusion layer. Finally, the fused features are input into another fully connected layer to realize the classification and recognition of the samples, and the network recognition result of PRI modulation type is obtained.

6. The non-ideal observation radar PRI modulation identification method according to claim 1, characterized in that: The process of determining the continuity parameters of the PRI pulse sequence is as follows: Assume that the continuity threshold of each PRI pulse sequence is: Among them, x i represents the i-th PRI pulse in the PRI pulse sequence, and the length of the PRI pulse sequence is N; The continuity parameter C is defined as the ratio of the number of PRI pulses M that meet the continuity requirement to the total length N of the PRI pulse sequence; the number of PRI pulses M that meet the continuity requirement is x i with x i-1 、x i+1 The number of pulses whose absolute value of the difference is lower than the continuity threshold ε.

7. The non-ideal observation radar PRI modulation identification method according to claim 1, characterized in that: The prior parameter interval of each PRI modulation type is determined in advance by statistical calculation. Then, each PRI pulse sequence is subjected to continuity parameter calculation. When its continuity parameter falls within the prior parameter interval of a certain PRI modulation type, the PRI pulse sequence is considered to be the PRI modulation type corresponding to this interval and is used as the continuity recognition result.

8. The non-ideal observation radar PRI modulation identification method according to claim 1, characterized in that: The PRI modulation types include fixed PRI, jitter PRI, staggered PRI, group-variable PRI, sliding PRI, and sinusoidal PRI; wherein: The sine PRI parameter range is 100-3000μs, with a staggered number of 5-20, a sliding step of 10-100μs, and a period number of 2-20; the group variable PRI parameter range is 100-3000μs, with a staggered number of 3-10, and a period number of 10-20; the sliding PRI parameter range is 300-500μs, with a staggered number of 3-20, and a sliding step of 10-100μs; the jitter PRI parameter range is 100-3000μs, with a jitter range of 2-20%; the staggered PRI parameter range is 100-3000μs, with a staggered number of 3-20; and the fixed PRI parameter range is 100-3000μs.

9. A terminal device comprising a processor, a memory, and a computer program stored in the memory; characterized in that: When the processor executes the computer program, the non-ideal observation radar PRI modulation identification method according to any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium storing a computer program; wherein: When the computer program is executed by a processor, the method for identifying PRI modulation of a non-ideal observation radar according to any one of claims 1 to 8 is implemented.

Images