A radar dynamic multi-feature data fusion target classification method

Abstract

The present invention relates to a target classification method based on dynamic multi-feature data fusion, belonging to the field of radar signal processing technology. The method comprises the following steps: 1) radar collects echo data of the target and pre-processes the echo data, and constructs a training data set using the signal's time-frequency information, distance period information, and distance period data; 2) constructs a dynamic multi-feature fusion network target classification model, including a global feature extraction module, a local feature extraction module, a dynamic feature extraction module, and a feature fusion module; 3) inputs the training data set to iteratively optimize and train the dynamic multi-feature fusion network to obtain optimal network parameters; 4) pre-processes the real-time radar echo signal, inputs the trained dynamic multi-feature fusion network for testing, and completes target classification. The present invention can simultaneously extract features of the radar signal's Doppler information and dynamic distance amplitude information, thereby improving the accuracy of target classification and meeting practical application requirements.

Description

Technical Field

[0001] The present invention relates to a target classification method based on dynamic multi-feature data fusion, which is particularly suitable for radar target classification intelligent processing and belongs to the technical field of radar signal processing. Background Art

[0002] Radar target detection and classification are widely used in both military and civilian applications. However, due to the impact of clutter and noise generated by complex environments and the diverse nature of target types, reliable and robust radar target detection and classification remain key technologies in need of research. Currently, the main challenges in radar target detection and recognition lie in background suppression, high-resolution target feature extraction, and complex feature classification. In recent years, small targets such as drones have gained popularity due to their small size, low price, and ease of control. They have been widely used in aerial photography, environmental monitoring, mobile communications, and other fields. However, the recognition and classification of small targets present certain challenges.

[0003] Several FMCW radar-based methods have been applied to aerial target recognition, leveraging radar cross-section (RCS) features to provide useful information for aerial target classification. However, when two targets are similar in size or small, RCS cannot reliably distinguish between them. Furthermore, RCS is affected by the scatterer's physical material, size, and shape, reducing target classification accuracy. Micro-Doppler, a Doppler shift caused by tiny components other than the target itself, was first used for aerial target recognition by Chen et al. (V.C. Chen, F. Li, S.-S. Ho, and H. Wechsler, "Micro-Doppler effect in radar: Phenomenon, model, and simulation study," IEEE Trans. Aerosp. Electron. Syst., vol. 42, no. 1, pp. 2–21, Jan. 2006). The rotation of drone rotors and the flapping of bird wings generate Doppler signals in radar echoes. Micro-motion characteristics are closely related to type and motion state, making micro-Doppler an effective feature for drone and bird classification. When analyzing micro-Doppler, the frequency of the spectrum is an important factor because the micro-motion of the target is repeatedly represented in the spectrum. By performing time-frequency transformation on the micro-motion signal, the rotation frequency and blade length of the drone blades are estimated, thereby distinguishing the type of target. However, due to the influence of time-frequency resolution, there is aliasing in the drone rotor in actual measurement, and it is difficult to clearly observe the flicker of each blade, which affects the classification of the target. In order to further analyze the micro-motion characteristics, the CVD graph is introduced. The CVD is obtained by iterative Fourier transform of each row of the spectrum. As an important feature for classification, considering that MDS is a Doppler signal in the time domain and CVD is a Doppler signal in the frequency domain, the CVD method is also effective for analyzing radar echo signals.

[0004] In recent years, numerous DNN-based radar automatic target recognition methods have been developed. Deep learning-based methods extract high-dimensional features through deep neural networks, avoiding the limitations of manual feature extraction. Xu et al. (Xu B, Chen B, Wan J, et al. Target-Aware Recurrent Attentional Network for Radar HRRP Target Recognition. Signal Processing 2019;155:268-80.) proposed a target-aware recurrent attention network (TARAN), which leverages the temporal dependencies between range units to recognize planar targets and a reusable long short-term memory (RLSTM) network to extract spatial features of gestures. For three-dimensional data analysis, a 3D-CNN and a hybrid complex architecture 3D-CNN-LSTM were proposed and experimented on radar data collected by ultra-wideband radar (UWB). Rahman et al. trained the GoogleNet framework to classify drones and birds based on their micro-Doppler characteristics. Kim et al. (Kim BK, Kang HS, Park S O. Drone classification using convolutional neural networks with merged Doppler images [J]. IEEE Geoscience and Remote Sensing Letters, 2016, 14 (1): 38-42.) merged micro-Doppler features and CVD features into a new image and improved the classification accuracy by using CNN structure for classification. Chen et al. (Chen X, Zhang H, Song J, et al. Micro-motion classification of flying bird and rotor drones via data augmentation and modified multi-scale cnn [J]. Remote Sensing, 2022, 14 (5): 1107.) used FMCW radar to collect data of drones and flying birds, fused the range periodogram and micro-Doppler features, and used multi-channel DCNN to classify targets, which improved the classification accuracy of drones and flying birds. It can be seen that using deep learning to extract target micro-Doppler features is very helpful for target classification and recognition. However, extracting features from a spectrum has certain limitations and is easily affected by complex environments, which reduces the accuracy of target classification. Summary of the Invention

[0005] In response to the shortcomings of the above-mentioned prior art, the present invention provides a radar dynamic multi-feature data fusion target classification method, which can simultaneously extract features of the Doppler information and dynamic range amplitude information of the radar signal, improve the accuracy of target classification, and meet practical application needs.

[0006] The present invention provides a radar dynamic multi-feature data fusion target classification method, which is special in that it includes the following steps:

[0007] Step 1: The radar collects the target's echo data and preprocesses the echo data, using the signal's time-frequency information, range-period information, and range-period data to construct a training dataset;

[0008] The echo data preprocessing method described in step 1 is as follows: collecting radar echo data under various observation conditions and areas, preprocessing the data, including removing negative frequencies and the DC component of the signal, and performing MTI processing on the signal to effectively filter out stationary clutter on the ground and in the air. After the above data preprocessing steps, a range period diagram of the target is drawn, the radar signal time series is intercepted according to the set sample observation time, time-frequency analysis is performed on each signal sample sequence, the time-frequency analysis results are Fourier transformed along the time dimension to obtain a CFD spectrum, and the time-frequency spectrum diagram and the CFD spectrum are spliced ​​into a TCD spectrum.

[0009] Preferably, the specific steps of the echo data preprocessing method described in step 1 are:

[0010] First, the radar echo signal is processed to remove negative frequency modulation, DC signal and MTI. Then, the target range period information is obtained. The target range unit information is extracted and the echo is converted into a time-frequency spectrum using the Short-time Fourier Transform (STFT). Assuming s(t) is the extracted target echo signal, the Short-time Fourier Transform formula can be expressed as:

[0011]

[0012] Where g(·) represents a sliding window, taking a Gaussian window function, t represents the time dimension, and ω represents the frequency dimension. By observing the range Doppler of the UAV, the target distance unit is selected, and the signal of the target distance unit is extracted and STFT is performed to obtain the time-frequency diagram of the target signal. The data matrix STFT is obtained by performing short-time Fourier transform on the preprocessed echo data, and the time-frequency transform result is FFT along the time dimension to obtain the CFD spectrum, that is,

[0013] CFD=F t {STFT} (2)

[0014] F tIndicates that FFT is performed on the time-frequency graph along the time dimension;

[0015] The time-frequency spectrum and CFD spectrum are horizontally spliced ​​into a TCD spectrum, which is expressed as

[0016] y tcd =[x md ,x cfd ] (3)

[0017] where y tcd Represents the TCD spectrum matrix after splicing, x md represents the micro-Doppler spectrum matrix, x cfd Represents the CFD spectrum matrix.

[0018] Preferably, the method for constructing the training data set in step 1 is:

[0019] The training dataset contains the range periodogram, time-frequency spectrum, and range-period matrix of multiple target signals, along with corresponding labels. A training sample is constructed based on the target's location information. The range unit of the target echo in the radar signal at each moment is determined, and the range unit data is extracted. Time-frequency analysis is performed on the data to obtain a time-frequency spectrum. The results of the time-frequency analysis are Fourier transformed along the time dimension to obtain a CFD spectrum. The two spectra are horizontally spliced ​​into one to construct the second training sample. To capture the dynamic characteristics of the data, a multi-frame range-period data cube is constructed and stored in a matrix:

[0020]

[0021] in Represents multi-frame periodic cube data, FFT represents Fourier transform along the time dimension, s IF (t m ,t s ) represents the intermediate frequency signal, t m represents the fast time of the time index within the mth chirp, t s represents the slow time between adjacent chirps;

[0022] Step 2: Construct a dynamic multi-feature fusion network target classification model, including a global feature extraction module, a local feature extraction module, a dynamic feature extraction module, and a feature fusion module;

[0023] Preferably, the dynamic multi-feature fusion network in step 2 includes an input layer, channel 1, channel 2 and channel 3, the input layer extracts preliminary features by inputting three data sets, channel 1 connects multiple local feature extraction modules to extract local features in the target spectrogram, channel 2 is a ResNet3D network model to extract the dynamic distance information of the target, and channel 3 is composed of multiple global feature extraction modules to extract global features in the target signal spectrogram; the feature fusion network further extracts local features and global features through spatial attention and channel attention, and finally adds the features extracted by the three channels. The number of nodes in the output layer of the feature fusion network is 6, indicating six categories; assuming that the inputs x1, x2, and x3 respectively represent the input time-frequency spectrum, the distance periodogram and the distance period data are respectively input into the network modules of the three channels, flattened in the channel direction, and then the number of channels of each pixel is linearly transformed, and finally the features extracted by the three channels are input into the feature fusion module;

[0024] Step 3: Input the training data set to iteratively optimize the dynamic multi-feature fusion network to obtain the optimal network parameters;

[0025] Preferably, the specific method of step 3 is:

[0026] Input the training data set constructed in step 1 into the target classification model established in step 2, and use the improved gradient descent method AdamW to train the model. After continuous iterative optimization training, when the label detected by the network is close to the set label, the target classification model training is completed and the network parameters are optimized.

[0027]

[0028] m t ←β1m t-1 (1-β1)g t (6)

[0029] v t ←β2v t-1 +(1-β2)g 2 t (7)

[0030]

[0031] Where t represents the number of current iterations and η represents the learning rate.

[0032] Step 4: Preprocess the real-time radar echo signal and input it into the trained dynamic multi-feature fusion network for testing to complete target classification.

[0033] Preferably, the specific method of step 4 is:

[0034] The real-time generated target echo data is preprocessed, and the time-frequency information and distance information obtained after processing are made into a time-frequency spectrum, a distance periodogram, and a distance data tensor in the form of an input data set. These information is then input into an optimal target detection model trained based on a large amount of high-quality data. After feature extraction and feature fusion, the output value is obtained to achieve target classification.

[0035] The output vector z of the neural network is [z1,z2,....,z6], where z i is the score of the model on the i-th category, and the Softmax function calculates the probability p of the i-th category i The formula is as follows:

[0036]

[0037] where z i is the score of the i-th category, p i is the predicted probability of class i.

[0038] The radar dynamic multi-feature data fusion target classification method of the present invention has the following beneficial effects:

[0039] (1) Breaking through the limitations of traditional classification methods, it can achieve intelligent classification of radar targets in complex environments;

[0040] (2) The present invention overcomes the shortcomings of single-feature neural network feature extraction, and simultaneously extracts and fuses the time-frequency information, distance information, and dynamic information in the radar echo signal to improve target classification performance;

[0041] (3) The present invention enhances the feature extraction capability of the global feature extraction module by utilizing the attention mechanism, uses channel attention and spatial attention in the feature fusion module to further extract local features and local features, and then fuses these features to improve the accuracy of target classification and meet practical application needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a flowchart of the target classification method;

[0043] Figure 2 It is a dynamic multi-feature fusion network framework diagram;

[0044] Figure 3 This is the structure diagram of the global feature extraction module;

[0045] Figure 4 It is the structural diagram of the local feature extraction module;

[0046] Figure 5 This is the structural diagram of the dynamic feature extraction module;

[0047] Figure 6 This is the structure diagram of the feature fusion module. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0049] A radar dynamic multi-feature data fusion target classification method of this embodiment includes the following steps:

[0050] 1. Radar signal preprocessing and data set construction

[0051] Radar signal preprocessing:

[0052] First, to eliminate negative frequency signals in the radar echo data, the data was preliminarily processed. A high-pass filter was applied to effectively remove the DC component in the signal. This step aims to enhance the dynamic range of the signal and ensure the accurate extraction of high-frequency components. Moving target indication (MTI) technology was used to effectively filter out stationary clutter on the ground and in the air, thereby significantly improving the target detection capability. By comparing the phase changes between adjacent pulses, MTI can effectively suppress stationary or slow-moving background noise, making the target signal more prominent. After the above data preprocessing steps, the target's range period diagram was drawn to clarify the range unit where the target is located. By analyzing the range period diagram, the target's echo data can be accurately extracted. In order to further extract the target's micro-motion characteristics, short-time Fourier transform (STFT) is used for time-frequency analysis. After the radar signal is preprocessed to remove jitter interference, the range unit where the target is located is found, and the echo signal of the target's range unit is extracted for STFT. The signal within the time window is subjected to FFT using a sliding window method to obtain a time-frequency spectrum diagram:

[0053]

[0054] Here, g(·) represents a sliding window, which in this paper uses a Gaussian window function. t represents the time dimension, and w represents the frequency dimension. By observing the UAV's range Doppler, we select the target's range cell. We extract the signal from this range cell and perform STFT on it to obtain a time-frequency plot of the target signal.

[0055] The data matrix STFT is obtained by performing short-time Fourier transform on the preprocessed echo data, and the CFD spectrum is obtained by performing FFT on the time-frequency diagram along the time axis.

[0056] CFD=F t {STFT} (2)

[0057] F tIndicates that FFT is performed on the time-frequency graph along the time axis.

[0058] The time-frequency spectrum and CFD spectrum are horizontally spliced ​​into a TCD spectrum, which is expressed as

[0059] y tcd =[x md ,x cfd ] (3)

[0060] where y tcd Represents the TCD spectrum matrix after splicing, x md represents the micro-Doppler spectrum matrix, x cfd Represents the CFD spectrum matrix.

[0061] In order to obtain the dynamic characteristics of the data, a multi-frame distance period data cube is constructed, and the dynamic characteristics between multiple frames are extracted through a three-dimensional network to improve the classification accuracy of the target. After preprocessing the collected original echo, the distance period data R is obtained by fast Fourier transforming the intermediate frequency signal along the fast time index. It is an N×M covariance matrix, where N represents the distance sampling unit and M represents the number of cycles. Due to the distance offset caused by the movement of the target, in order to extract the dynamic characteristics of the target, L frames of the distance period sequence can be obtained within a total imaging time, and then the multi-frame distance period sequence plane is superimposed along the slow time axis to form the distance period sequence tensor data. Where N represents the distance sampling unit, M represents the number of cycles, and L represents the number of frames.

[0062]

[0063] Since each sample is input into the network for processing multiple times during the training process, in order to reduce the number of time-frequency transformation operations, the data is first preprocessed and the preprocessed data is used as the data set.

[0064] 2. Dataset Construction

[0065] After data preprocessing, the target range periodogram, time-frequency spectrum, and TCD spectrum are obtained, which are then constructed into three datasets. The training dataset contains the range periodogram, time-frequency spectrum, and range-period matrix of multiple signals, along with their corresponding labels. A training sample is constructed based on the target's location information. The range unit of the target echo in the radar signal at each moment is determined, and the data for this range unit is extracted. Time-frequency analysis is performed on the data to obtain a time-frequency spectrum. The results of the time-frequency analysis are Fourier transformed along the time dimension to obtain a CFD spectrum. The two spectra are horizontally concatenated into one to construct the second training sample. To capture the dynamic characteristics of the data, a multi-frame range-period data cube is constructed and stored in a matrix.

[0066] A dynamic multi-feature fusion convolutional neural network classification method includes the following steps:

[0067] Build a dynamic multi-feature fusion network.

[0068] Compare with Figure 2 ,The constructed classification network model structure is divided into three parts: global ,feature extraction module, local feature extraction module, dynamic feature ,fusion module, and softmax classification.

[0069] (1) Global feature extraction module

[0070] In the global feature extraction module channel, the first step is Patch Partition, which divides the input (H, W, 3) image into (4, 4) small blocks. The size of the block image is (H / 4, W / 4, 48) dimensions. Linear Embedding maps the block image to 96 dimensions. Figure 3 As shown in the figure, Windows Multi-Head Self-Attention (W-MSA) is introduced in the global feature extraction branch. W-MSA is a module in the Swin-transformer. The computational complexity of its feature map extraction grows linearly with the size of the feature map and utilizes the prior knowledge of image locality. However, there is a lack of information interaction between windows. Therefore, the next module introduces SW-MSA. By moving the window toward the lower right corner, the pixels in different windows interact, better obtaining the context information and extracting the correlation information between each window.

[0071] Ω(MSA)=4hwC 2 +2(hw) 2 C (12)

[0072] Ω(W-MSA)=4hwC 2 +2M 2 hwC (13)

[0073] Where h represents the height of the feature map, w represents the width of the feature map, C represents the depth of the feature map, and M represents the size of each window. For each stage, by incorporating the patch into the global feature block, the feature map passes through the LayerNorm layer into the W-MSA, and then passes through the linear layer with the GELU activation function, as shown in (14).

[0074] G i =f(SW-MSA(LN(f(W-MSA(LN(G i-1 ))))+G i-1 ))+f(W-MSA(LN(G i-1 )))+G i-1 (14)

[0075] Among them G i Represents the output of the global feature, f is the convolution operation with a convolution kernel size of 1×1, LN represents the LayerNorm operation, and finally, the global features in the extracted time-spectrogram are input into the feature fusion module.

[0076] (2) Local feature extraction module

[0077] For the local feature extraction module, we hope that the network can only extract local features. By borrowing the LN and GELU activation functions in Transform, we can achieve good performance in different scenarios. Specifically, the local feature extraction module. The local features extracted from the distance periodogram are input to the feature fusion module. This process is shown in (15).

[0078] L i =f(f(LN(f 7×7 (L i-1 ))))+L i-1 (15)

[0079] Among them L i is the local feature extracted, f represents the convolution operation with a convolution kernel size of 1×1, LN represents the LayerNorm module, and f 7×7 Indicates a convolution operation with a convolution kernel of 7×7.

[0080] (3) Dynamic feature extraction module

[0081] ResNet3D has a specific structure similar to ResNet, but its convolution is extended to three-dimensional convolution to process tensor data. Specifically, the input data is confined to (batch-size, channels, depth, height, width), and through a series of 3D convolution layers, 3D pooling layers and residual connections, a feature representation for classification or regression is finally obtained. The input of the network is a 16×513×64 matrix, which divides the continuous distance period data into 16 frames, which enables the network to learn rich spatial features, which is crucial for accurately recognizing complex data. The three-dimensional convolution kernel slides on the distance period sequence tensor data, which can capture the long-term dependencies and related features between different numbers of frames in the sequence, and performs convolution operations on the dataset to obtain a three-dimensional feature map. The structure of ResNet3D includes 4 3×3×3, 64-channel convolutions, 4 3×3×3, 128-channel convolutions, 4 3×3×3, 256-channel convolutions, and 4 3×3×3, 512-channel convolutions, two of which have a residual connection.

[0082] (4) Feature fusion module

[0083] In the feature fusion module, global features are fed into the CBAM module, which exploits the interdependencies between channel maps to refine the feature representations for specific semantics. Local features are then fed into the spatial attention (SA) mechanism to enhance local details and suppress irrelevant regions. Finally, the results from each attention and fusion path are fused and connected to a residual inverted MLP. This prevents problems such as vanishing gradients, exploding gradients, and network degradation to a certain extent, effectively capturing global and local feature information at all levels.

[0084] CA(x)=σ(MLP(AvgPool(x))+MLP(MaxPool(x))) (16)

[0085] SA(x)=σ(f 7×7 (Concat[AvgPool(x),MaxPool(x)])) (17)

[0086] Among them, σ represents the Sigmoid function, f 1×1 Indicates a convolution operation with a convolution kernel of 1×1, f 7×7 The convolution operation with a convolution kernel of 7×7 is represented by the feature fusion module output as follows:

[0087]

[0088] in represents element-wise multiplication, represents the data after channel attention, represents the data after spatial attention, Represents the data output after feature fusion.

[0089] The specific steps of step 3 are:

[0090] The AdamW gradient descent method is used in iterative optimization training. Given the initial learning rate parameter α, momentum factors β1, β2, and optimization parameters θ t , the first instantaneous vector m t , the second instantaneous vector v t , progress multiplier η t

[0091]

[0092] m t ←β1m t-1 +(1-β1)g t (6)

[0093] v t ←β2v t-1 +(1-β2)g 2t (7)

[0094]

[0095] Where t represents the number of current iterations, and η represents the learning rate;

[0096] Finally, after the average pooling layer and the fully connected layer, the probability p of the i-th class is calculated by the Softmax function i The formula is as follows:

[0097]

[0098] where z i is the score of the i-th category, p i is the predicted probability of class i.

Claims

1. A radar dynamic multi-feature data fusion target classification method, characterized by The following steps are involved: Step 1: The radar collects the target's echo data and preprocesses the echo data, using the signal's time-frequency information, range-period information, and range-period data to construct a training dataset; Step 2: Construct a dynamic multi-feature fusion network target classification model, including a global feature extraction module, a local feature extraction module, a dynamic feature extraction module, and a feature fusion module; Step 3: Input the training data set to iteratively optimize the dynamic multi-feature fusion network to obtain the optimal network parameters; Step 4: Preprocess the real-time radar echo signal and input it into the trained dynamic multi-feature fusion network for testing to complete target classification; The method for constructing the training dataset described in step 1 is: The training dataset contains the range periodogram, time-frequency spectrum, and range-period matrix of multiple target signals, along with corresponding labels. A training sample is constructed based on the target's location information. The range unit of the target echo in the radar signal at each moment is determined, and the range unit data is extracted. Time-frequency analysis is performed on the data to obtain a time-frequency spectrum. The results of the time-frequency analysis are Fourier transformed along the time dimension to obtain a CFD spectrum. The two spectra are horizontally spliced ​​into one to construct the second training sample. To capture the dynamic characteristics of the data, a multi-frame range-period data cube is constructed and stored in a matrix: in Represents multi-frame periodic cube data, FFT represents Fourier transform along the time dimension, s IF (t m ,t s ) represents the intermediate frequency signal, t m represents the fast time of the time index within the mth chirp, t s represents the slow time between adjacent chirps; The dynamic multi-feature fusion network in step 2 includes an input layer, channel 1, channel 2, and channel 3. The input layer extracts preliminary features by inputting three data sets. Channel 1 connects multiple local feature extraction modules to extract local features in the target spectrogram. Channel 2 is a ResNet3D network model that extracts dynamic distance information of the target. Channel 3 is composed of multiple global feature extraction modules to extract global features in the target signal spectrogram. feature The fusion network will further extract local features and global features through spatial attention and channel attention, and finally add the features extracted from the three channels. The number of nodes in the output layer of the feature fusion network is 6, representing six categories; assuming that the input x1, x2, and x3 represent the input time-frequency spectrum, the distance periodogram and distance period data are input into the network modules of the three channels respectively, flattened in the channel direction, and then the number of channels of each pixel is linearly transformed. Finally, the features extracted from the three channels are input into the feature fusion module.

2. The radar dynamic multi-feature data fusion target classification method according to claim 1, characterized in that: The echo data preprocessing method described in step 1 is: collecting radar echo data under various observation conditions and areas, preprocessing the data, including removing negative frequencies and the DC component of the signal, and performing MTI processing on the signal to effectively filter out stationary clutter on the ground and in the air. After the above data preprocessing steps, a range period diagram of the target is drawn, the radar signal time series is intercepted according to the set sample observation time, time-frequency analysis is performed on each signal sample sequence, the time-frequency analysis results are Fourier transformed along the time dimension to obtain a CFD spectrum, and the time-frequency spectrum diagram and the CFD spectrum are spliced ​​into a TCD spectrum.

3. A radar dynamic multi-feature data fusion target classification method according to claim 2, characterized in that: The specific steps of the echo data preprocessing method described in step 1 are: First, the radar echo signal is processed by removing negative frequency modulation, DC signal and MTI, and then the target range period information is obtained. The target range unit information is extracted and the echo is converted into a time-frequency spectrum using Fourier transform (STFT). Assume that s(t) is the extracted target echo signal, where the short-time Fourier transform formula is expressed as: Where g(·) represents a sliding window, taking a Gaussian window function, t represents the time dimension, and ω represents the frequency dimension. By observing the range Doppler of the UAV, the target distance unit is selected, and the signal of the target distance unit is extracted and STFT is performed to obtain the time-frequency diagram of the target signal. The data matrix STFT is obtained by performing short-time Fourier transform on the preprocessed echo data, and the time-frequency transform result is FFT along the time dimension to obtain the CFD spectrum, that is, CFD=F t {STFT} (2) F t Indicates that FFT is performed on the time-frequency graph along the time dimension; The time-frequency spectrum and CFD spectrum are horizontally spliced ​​into a TCD spectrum, which is expressed as y tcd =[x md ,x cfd ] (3) where y tcd Represents the TCD spectrum matrix after splicing, x md represents the micro-Doppler spectrum matrix, x cfd Represents the CFD spectrum matrix.

4. The radar dynamic multi-feature data fusion target classification method according to claim 1, characterized in that: The specific method of step 3 is: The training data set constructed in step 1 is input into the target classification model established in step 2, and the model is trained using the improved gradient descent method AdamW. After continuous iterative optimization training, the target classification model training is completed and the network parameters are optimized until the labels detected by the network are close to the set labels.

5. The radar dynamic multi-feature data fusion target classification method according to claim 1, characterized in that: The specific method of step 4 is: The real-time generated target echo data is preprocessed, and the time-frequency information and distance information obtained after processing are made into a time-frequency spectrum, a distance periodogram, and a distance data tensor in the form of an input data set. These information is then input into an optimal target detection model trained based on a large amount of high-quality data. After feature extraction and feature fusion, the output value is obtained to achieve target classification. The output vector z of the neural network is [z1,z2,....,z6], where z i is the score of the model on the i-th category, and the Softmax function calculates the probability p of the i-th category i The formula is as follows: where z i is the score of the i-th category, p i is the predicted probability of class i.

Images