Radar active composite interference signal identification and parameter measurement method of Yolov8 network based on convolution attention mechanism and rejection loss

Abstract

The invention discloses a radar active composite interference signal identification and parameter measurement method of a Yolov8 network based on a convolution attention mechanism and rejection loss, and aims to solve the problems of single working mode and poor robustness of the existing interference signal identification method and convert the interference identification problem into the target detection problem. And a target detection framework is applied to detect, identify and measure parameters of interference signals, and a CBAM attention mechanism is introduced, so that the image feature extraction capability of the network is enhanced, and the identification rate is improved. In addition, a rejection loss function is also introduced, so that a good regression prediction effect can be kept when the interference signal time-frequency images are overlapped more frequently. The network provided by the invention has good detection and identification performance on most of signals under the condition of low interference-to-noise ratio. The invention belongs to the technical field of radar interference signal detection and identification.

Description

Technical Field

[0001] The present invention belongs to the technical field of radar interference signal detection and recognition, and specifically relates to a radar active composite interference signal recognition and parameter measurement method based on a Yolov8 network of a convolutional block attention module (CBAM) and a repulsion loss (REP). Background Art

[0002] With the increase in the number of different radiation sources on modern battlefields, the electromagnetic environment has become increasingly complex, with high signal density, high degree of pulse sequence interleaving, and fierce competition and confrontation on the spectrum. The radar will receive multiple interference signals, which are aliased in the time domain or frequency domain to form a composite interference signal. Since the received mixed signal may contain an unknown amount of interference, it will have a great impact on radar detection. In order to take more effective and targeted anti-interference measures, it is necessary to accurately and quickly detect the signal, identify the type, and measure the parameters of the interference signals that may exist in the radar echo, so as to ensure the working performance of the radar.

[0003] By searching the current relevant literature on compound interference identification, it was found that Qu, Qizhe proposed an interference classification network JRNet in "JRNet: Jamming recognition networks for radar compound suppression jamming signals" published in "IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY" (2020, 69(12): 15035-15045). The network uses power spectrum features as network input and can identify 4 single interference types and 6 compound interference types. Although this method has good recognition performance under low interference-to-noise ratio, it cannot perform multi-task processing of signal detection, type recognition, and parameter measurement on interference signals at the same time. JiaXiang Zhang published “Radar Compound Jamming Cognition Based on a Deep Object Detection Network” in “IEEE TRANSACTIONS ONAEROSPACE AND ELECTRONIC SYSTEMS” (2023, 59(3): 3251-3263). He applied a target detection network to identify the interference type based on the time-frequency image and used a prediction box to locate the position of the interference signal time-frequency image on the time-frequency image. He also gave a corresponding parameter estimation method. However, no experiments were conducted for low interference-to-noise ratios below 0 dB. Summary of the invention

[0004] In order to solve the problems of single working mode and poor robustness of existing interference signal recognition methods, the present invention proposes a radar active composite interference signal recognition and parameter measurement method based on Yolov8 network with convolutional attention mechanism and rejection loss.

[0005] The technical solution adopted by the present invention to solve the above problems is:

[0006] The present invention comprises the following steps:

[0007] Step 1, establishing a mathematical model of radar composite interference signal;

[0008] The input signal model is:

[0009]

[0010] In the formula, k represents the number of components that constitute a single interference signal, which is composed of at most q interference components, J i (n) represents the received interference signal, A i is the amplitude coefficient of the modulation when propagating through the environment, awgn(n) represents Gaussian white noise, x t (n) represents the real target echo signal.

[0011] Step 2, preprocessing the composite interference signal;

[0012] First, the composite interference signal is transformed into a time-frequency image X(t,ω), which is expressed as follows:

[0013]

[0014] Where ω represents frequency and h(τ-t) is the analysis window function.

[0015] Step 3: Input the training set into the Yolov8 network with convolutional attention mechanism and exclusion loss for training;

[0016] The Yolov8 network with convolutional attention mechanism and exclusion loss includes a context-aware cross-level fusion network module, a pyramid pooling module, a convolutional attention module, a splicing module and a detection head module, etc., wherein the classification branch uses a cross entropy loss function, and the regression branch uses an exclusion loss function and an intersection-over-union loss function.

[0017] Step 4: Input the training set video images into the Backbone part of the network for feature extraction, including:

[0018] Step 4.1: The input image is operated through a convolution module to extract features;

[0019] Step 4.2: Input the feature map obtained by the convolution layer into the context-aware cross-level fusion network module. The context-aware cross-level fusion network module includes a 1x 1 convolution layer and n BottleNeck layers.

[0020] Step 4.3: Input the above features into the pyramid pooling module. Pyramid pooling is improved from SPP in Yolov5. It consists of three maximum pooling layers, two convolutional layers and a feature fusion layer. It performs pooling operations of different sizes on the feature map and integrates the results to obtain a fixed-size output.

[0021] Step 5: Input the feature map into the Neck part of the network, perform upsampling, splicing and other operations to perform feature fusion, and introduce the convolutional attention module to enhance the feature extraction capability and improve the recognition rate.

[0022] Step 5.1, upsample the feature map so that the output size is twice the input size for easy subsequent processing;

[0023] Step 5.2, feature fusion is performed on the feature map obtained by upsampling and the feature map obtained by the context-aware cross-level fusion network module, and the feature maps of 80*80*256 and 80*80*512 are spliced ​​into a feature map of 80*80*768. The subsequent splicing modules are similar;

[0024] Step 5.3: After the spliced ​​feature map is passed through the context-aware cross-level fusion network module, it is input into the convolutional attention module, which consists of a channel attention module and a spatial attention module.

[0025] Step 6: Input the fused features of the Neck part into the detection head module, further extract the features through the convolution module and the two-dimensional convolution module, and change its dimension to facilitate the recognition operation and regression prediction box operation. In this part, the exclusion loss function is introduced to maintain a good regression prediction effect when the interference signal time-frequency image overlaps a lot. Use the cross entropy loss function, the exclusion loss, and the intersection-over-union loss function to continuously correct the prediction results of the category and the prediction box, and calculate the pulse width, bandwidth, center frequency and other parameters according to the coordinates of the prediction box.

[0026] Step 7: Input the test set into the trained Yolov8 network with convolutional attention mechanism and rejection loss to obtain the interference signal components and interference signal parameters contained in the input signal.

[0027] The beneficial effects of the present invention are:

[0028] 1. Compared with the prior art, the present invention designs a more intelligent composite interference signal detection and identification and parameter measurement method to solve the problems of single working mode and poor robustness of the prior method, converts the interference identification problem into a target detection problem, and applies the target detection framework to detect, identify and measure the parameters of the interference signal;

[0029] 2. The present invention conducts combined modeling analysis on multiple interference signals, designs a Yolov8 network with convolutional attention mechanism and exclusion loss from the perspective of signal time-frequency image, and introduces convolutional attention mechanism to enhance the network's ability to extract image features and improve recognition rate; in addition, the exclusion loss function is introduced to maintain good regression prediction effect even when the interference signal time-frequency image overlaps more;

[0030] 3. The present invention uses multiple modules such as context-aware cross-level fusion network, convolutional attention, pyramid pooling, etc. to extract features from the input time-frequency graph, solving the problem of single working mode in traditional methods;

[0031] 4. The present invention can simultaneously complete the tasks of detection and identification and parameter measurement, while maintaining good detection and identification performance under low interference-to-noise ratio. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is the overall flow chart of the present invention;

[0033] Figure 2 It is a schematic diagram of the structure of Yolov8 with convolutional attention mechanism and rejection loss of the present invention;

[0034] In the figure, the time-frequency diagram generated by the signal after time-frequency transformation is first input into the Backbone module of the network for feature extraction, and then the features are input into the Neck module for feature fusion, and finally the processed features are input into the detection head module for classification and regression.

[0035] Figure 3 is the performance change curve of the network with the number of iterations in the present invention;

[0036] The six signals of slice reconstruction interference, white noise convolution interference, white noise product interference, spectrum dispersion interference, intermittent sampling forwarding interference and echo signal are combined to form a composite interference signal data set. Each combination generates time-frequency images at 11 interference-to-noise ratios of -10db, -8dB, -6dB, -4dB, -2dB, 0dB, 2dB, 4dB, 6dB, 8dB, and 10dB, and a total of 10,400 time-frequency images are generated to form a data set. It can be seen from the training result diagram that the method designed by the present invention has a higher recognition performance, and the mAP50:95 reaches 0.94.

[0037] Figure 4 Detection and identification parameter measurement results for NCN, CI, and SMSP composite interference;

[0038] The labels on the upper side of the detection box in the figure are signal category, confidence, pulse width, bandwidth and center frequency respectively.

[0039] Figure 5 This is the curve of mAP50:95 changing with the interference-to-noise ratio JNR.

[0040] Depend on Figure 5 It can be seen that the model maintains good performance at all interference-to-noise ratios, with mAP50:95 reaching 0.943 at 10db and 0.85 at -10db. DETAILED DESCRIPTION

[0041] like Figure 1 As shown in FIG. 1 , it is the overall flow chart of the present invention. First, the composite interference signal is transformed into a time-frequency image and a data set is prepared. Then, the data set is divided into a training set and a test set. The network of this invention is trained with the training set, and the performance of the network is verified with the test set. Specifically, the following steps are included:

[0042] Step 1: Establish the mathematical model of radar composite interference signal.

[0043] The simulation generates radar composite jamming signals, which include radar echo signals, suppression jamming, smart noise jamming, deception jamming, etc. The input signal model is:

[0044]

[0045] In the formula, k represents the number of components constituting a single interference signal, J i (n) represents the received interference signal, A i is the amplitude coefficient of the modulation when propagating through the environment, awgn(n) represents Gaussian white noise, x t (n) represents the real target echo signal.

[0046] Step 2: Preprocess the composite interference signal.

[0047] The composite interference signal is transformed into a time-frequency image to obtain the time-frequency joint distribution image of the signal. The interference signal is transformed into a two-dimensional time-frequency image X(t,ω) by using short-time Fourier transform.

[0048]

[0049] Where ω represents the frequency and h(τ-t) is the analysis window function.

[0050] The time-frequency image is preprocessed and the image is scaled normalized to a 640*640 image.

[0051] Step 3: Input the training set into the Yolov8 network with convolutional attention mechanism and rejection loss for training.

[0052] Step 4: Input the image into the Backbone part of the network for feature extraction.

[0053] Step 4.1: The input image is operated through the convolution module to extract features. The convolution module includes a convolution layer, a normalization layer, and then activated with the SiLU function. Compared with the ReLU function, the SiLU function has a smoother curve when it is close to zero, and because it uses the sigmoid function, the output range of the network can be between 0 and 1. The expression of the SiLU function is:

[0054]

[0055] Step 4.2: Input the feature map obtained by the convolution layer into the context-aware cross-level fusion network module, which includes a 1x 1 convolution layer and n BottleNeck layers. First, the input feature map is transformed by the first convolution layer, and the output feature map is divided into two parts. These two parts are processed by different convolution layers and then merged. This process helps the model capture more context information and thus identify the target more accurately.

[0056] The role of the context-aware cross-level fusion network module is to improve the performance and accuracy of the model. By introducing the context-aware cross-level fusion network module, the model can better capture the complex features in the image, thereby achieving better results in the object detection task. In addition, the context-aware cross-level fusion network module also has good scalability, which can further improve the feature extraction capability of the model without significantly increasing the computational cost.

[0057] Step 4.3: Input the features obtained above into the pyramid pooling module. Pyramid pooling is improved from SPP in Yolov5. It consists of three maximum pooling layers, two convolutional layers and feature fusion layers. It performs pooling operations of different sizes on the feature map and integrates the results to obtain a fixed-size output. This technology can effectively handle targets with varying sizes, thereby improving the generalization ability and robustness of the neural network.

[0058] Step 5: Input the feature map into the Neck part of the network for feature fusion.

[0059] Step 5.1: Upsample the feature map so that the output size is twice the input size for easy subsequent processing.

[0060] Step 5.2: Fuse the feature map obtained by upsampling and the feature map obtained by the context-aware cross-level fusion network module, and splice the feature maps of 80*80*256 and 80*80*512 into a feature map of 80*80*768. The subsequent splicing modules are the same.

[0061] Step 5.3: After the spliced ​​feature map passes through the context-aware cross-level fusion network module, it is input into the convolutional attention module. The convolutional attention consists of a channel attention module and a spatial attention module.

[0062] In the channel attention module, two different pooling operations are performed on the input features of dimension H×W×C to obtain two channel information, which are then input into the multi-layer perceptron for dimensionality reduction to obtain two feature maps of dimension 1×1×C. After element-by-element summation and processing with the Sigmoid activation function, the channel attention features are obtained.

[0063] The feature map obtained by element-wise multiplication of the channel attention feature and the input feature map is input into the spatial attention module. The above two pooling operations are also performed to obtain two feature maps with dimensions of H×W×1. The two feature maps are connected and a 7×7 convolution operation is performed. The spatial attention feature is then obtained after being processed by the Sigmoid activation function.

[0064] Finally, the spatial attention feature is multiplied by the input feature of the spatial attention module to obtain the final generated feature. After the introduction of convolutional attention, the feature extraction ability of the network is enhanced, allowing the network to pay more attention to key information, and ultimately the recognition rate of the signal is higher.

[0065] Step 6: The results of the last three context-aware cross-level fusion network modules are input into the detection head module for classification and regression.

[0066] The detection head module consists of two layers, one for classification and the other for regression prediction. Each layer consists of two convolution modules and one two-dimensional convolution module. The cross entropy function is used as the loss function for the classifier. The exclusion loss function and the intersection-over-union loss function are used as the loss function for the regression part.

[0067] The expression of the rejection loss function is:

[0068] L=L Attr +α*L RepGT +β*L RepBox

[0069] Where, LAttr is the amount of attraction between the predicted box and its true target box, expressed as follows

[0070]

[0071] in,

[0072]

[0073] Where P + is the set of detection boxes P divided according to the IoU threshold, Smooth L1 is the SmoothL1 distance. B P Represents the given P∈P + The predicted box of regression, represents the ground-truth bounding box with the maximum intersection-over-union ratio with P.

[0074]

[0075] Where, L RepGT It is B P and The loss of the overlapping part between the predicted box P and the surrounding target box G is made as far away as possible. RepBox is the loss function for the overlapping parts of the boxes of different targets i, j, so that different predicted boxes P i and P j Stay as far away as possible.

[0076] The IoU metric is a common metric in target detection. The intersection over union loss function is defined as the negative logarithm of the ratio of the intersection and union between the predicted box and the true box, which can measure the difference between the predicted box and the true box. Assume that the coordinates of the true box are The coordinates of the prediction box are x=(x t ,x b ,x l ,x r ).

[0077]

[0078] Step 7: Obtain the various categories of the input composite interference signal and its position on the time-frequency image, and calculate the signal parameters according to the prediction box.

[0079] mAP50:95 is used as the performance indicator to evaluate the model, where mAP (MeanAverage Precision) is the mean average precision. Based on the accuracy-recall curve, the average precision AP (Average Precision) can be obtained by calculating the average value of the accuracy value corresponding to each recall value. The expression of AP is as follows:

[0080]

[0081] where r i is the recall value corresponding to the i-th precision interpolation segment arranged in ascending order.

[0082] mAP calculates the AP of all categories and then takes the average value, which can measure the detection ability of the model in all categories.

[0083]

[0084] In the formula, K is the number of types.

[0085] mAP50:95 is the mean of mAP for all confidence levels between 50 and 95 with a step size of 5.

[0086] The above is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technician familiar with this profession can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent replacement and improvement made to the above embodiments without departing from the content of the technical solution of the present invention, based on the technical essence of the present invention, within the spirit and principles of the present invention, still fall within the protection scope of the technical solution of the present invention.

Claims

1. Radar active composite jamming signal recognition and parameter measurement method based on Yolov8 network with convolutional attention mechanism and rejection loss, characterized in that: The steps include: Step 1, establishing a mathematical model of radar composite interference signal; Step 2, preprocessing the composite interference signal; Step 3: Input the training set into the Yolov8 network with convolutional attention mechanism and exclusion loss for training; Step 4: Input the training set video images into the backbone part of the network for feature extraction; Step 5: Input the feature map into the Neck part of the network for feature fusion; Step 6: Input the fused features of the Neck part into the detection head module for classification and regression; Step 7: Input the test set into the trained Yolov8 network with convolutional attention mechanism and rejection loss to obtain the interference signal components and interference signal parameters contained in the input signal.

2. The radar active composite interference signal identification and parameter measurement method of the Yolov8 network based on the convolutional attention mechanism and the rejection loss according to claim 1 is characterized in that: In step 1, the input signal model is: In the formula, k represents the number of components constituting a single interference signal, J i (n) represents the received interference signal, A i is the amplitude coefficient of the modulation when propagating through the environment, awgn(n) represents Gaussian white noise, x t (n) represents the real target echo signal.

3. The radar active composite interference signal identification and parameter measurement method of the Yolov8 network based on the convolutional attention mechanism and the rejection loss according to claim 1 is characterized in that, In step 2, the composite interference signal is transformed into a time-frequency image X(t,ω), which is expressed as follows: Where ω represents frequency and h(τ-t) is the analysis window function.

4. The radar active composite interference signal identification and parameter measurement method of the Yolov8 network based on the convolutional attention mechanism and the rejection loss according to claim 1 is characterized in that, Step 4 includes: Step 4.1: The input image is operated through a convolution module to extract features; Step 4.2: Input the feature map obtained by the convolution module layer into the context-aware cross-level fusion network module. The context-aware cross-level fusion network module includes a 1x 1 convolution module layer and n BottleNeck layers. Step 4.3: Input the above features into the pyramid pooling module. Pyramid pooling is improved from SPP in Yolov5. It consists of three maximum pooling layers, two convolutional layers and a feature fusion layer. It performs pooling operations of different sizes on the feature map and integrates the results to obtain a fixed-size output.

5. The radar active composite interference signal identification and parameter measurement method of the Yolov8 network based on the convolutional attention mechanism and the rejection loss according to claim 1 is characterized in that, Step 5 includes: Step 5.1, upsample the feature map so that the output size is twice the input size for easy subsequent processing; Step 5.2, feature fusion is performed on the feature map obtained by upsampling and the feature map obtained by the context-aware cross-level fusion network module, and the feature maps of 80*80*256 and 80*80*512 are spliced ​​into a feature map of 80*80*768. The subsequent splicing modules are similar; Step 5.3: After the spliced ​​feature map is passed through the context-aware cross-level fusion network module, it is input into the convolutional attention module, which consists of a channel attention module and a spatial attention module.

Images