1289 results about "Radar signal processing" patented technology

The invention relates to the technical field of radar signal processing, in particular to a radar target analytical calculation method based on multi-dimensional data fusion and a radar device. Comprising the following steps: deploying a multi-band radar sensor array comprising an X band, a C band and a Ku band in a radar monitoring area; performing pulse compression and Doppler processing on the time domain echo signal, and extracting a time domain feature; spectral analysis is carried out on the frequency domain signals, and frequency domain features are extracted; performing angle estimation on the spatial signals, and extracting spatial features; a dynamic weight distribution model is constructed, a fusion weight is calculated through an adaptive algorithm based on three-dimensional quality indexes of a real-time signal-to-noise ratio (SNR), feature stability (SI) and data integrity (CI), and a joint representation vector containing time domain, frequency domain and space multi-dimensional information is generated. According to the invention, by deploying the multi-band radar sensor array, the recognition capability of the subtle feature difference of the target is improved.

The invention provides a ship radar identification method based on deep learning, and relates to the technical field of ship radar signal processing. The method comprises the following steps: denoising and filtering echo signals reflected by a target, and generating spatial features and filtered target feature signals based on the denoised and filtered signals; carrying out enhancement and normalization processing on the filtered target characteristic signal, and further carrying out space-time decomposition and nonlinear fusion to obtain a fused space-time signal matrix; windowing processing is carried out on the fused space-time signal matrix, and space-time feature optimization is carried out; generating data after spatio-temporal feature fusion through a spatio-temporal feature adaptive optimization algorithm; and inputting the data after spatio-temporal feature fusion into the target recognition model, outputting a prediction result and a confidence value of the target category, and optimizing the target recognition model. The problems that a traditional radar target recognition method is difficult to effectively extract accurate spatio-temporal features and lacks adaptive ability, and the generalization ability of a target recognition model is insufficient are solved.

The invention discloses a radar multi-angle multi-feature fusion target classification method, and relates to the technical field of radar signal processing, and the method comprises the steps: obtaining a multi-angle multi-feature fusion network target classification model; the multi-angle multi-feature fusion network target classification model comprises a time sequence feature extraction module, an attention fusion module and a feature fusion module; acquiring radar echo data of a target in real time; processing the radar echo data of the target to obtain processed radar echo data; the processed radar echo data comprises a time domain signal component graph of the target and a time frequency graph of the target; and inputting the processed radar echo data into a multi-angle multi-feature fusion network target classification model to obtain a target classification result. According to the invention, the accuracy of target classification can be improved.

The invention belongs to the technical field of radar signal processing and target identification, and particularly relates to a maritime hybrid target identification method and system based on modal decomposition and reconstruction, and the method comprises the steps: carrying out the variational modal decomposition of a radar echo signal of a maritime hybrid target, and decomposing an original signal into a plurality of intrinsic modal signals; estimating a background noise energy reference and removing noise modals based on the decomposed modal signals, clustering the remaining modals according to the center frequency, and combining the modals with the frequency within the center frequency range into an independent single-target signal which is of the same target and is reconstructed into an independent single-target signal; respectively carrying out time-frequency analysis on each reconstructed single target signal to obtain a corresponding time-frequency diagram, and extracting time-frequency domain features from the time-frequency diagram; and inputting the extracted features into a trained support vector machine classifier to realize automatic identification of the ship target and the floating target. A ship target and a floating target can be distinguished more accurately in a mixture scene, and the stability of tracking identification is improved.

The invention relates to a radar PRI modulation type identification method based on a neural network and a multi-head attention mechanism, and belongs to the field of radar signal processing. First-order difference is carried out on a radar pulse TOA sequence to obtain sample data of a PRI sequence, a median filter is adopted to carry out filtering processing on the sample data to reduce the influence of missing or false pulses, and a model integrating a multi-scale convolution extrusion excitation neural network, a bidirectional gating circulation unit neural network and a multi-head attention mechanism is constructed. The model can extract data local refinement features and capture data long-term time sequence correlation features, a multi-head attention mechanism is used for integrating global information, more weights are applied to important time steps in a gating circulation unit, the model is trained and tested, and the result shows that compared with other neural network methods, the model has the advantages that the algorithm is simple, and the algorithm is easy to implement. According to the method, the recognition rate of six kinds of radar PRI modulation signals is higher under the condition that the ratio of missing pulses to stray pulses is high.

The invention belongs to the technical field of radar signal processing, and discloses a radar scattering rapid test method based on polarization decomposition. Comprising the following steps: scanning a target 3D model, simulating scattered field distribution in different incident angles and polarization modes, calculating polarization contrast, and identifying a polarization sensitive area; establishing a dynamic coordinate system conversion model, fusing RTK data, IMU data and laser tracking data, and correcting the position of the trolley and the model in combination with a Doppler frequency shift model and a least square method; calculating an optimal test path of the trolley according to the sensitive area by adopting a natural heuristic optimization algorithm; establishing a geometric environment model, obtaining a back scattering signal through beam forming, analyzing a polarization coherence matrix, obtaining a high-entropy region, and decomposing scattering components; according to the method, closed-loop optimization is formed from a data acquisition source to a scattering component analysis tail end, and the test efficiency and the real-time performance are remarkably improved.

The invention provides a sleep monitoring model training method, a sleep monitoring method and equipment. The method is applied to radar signal processing. The method comprises the following steps: acquiring a data set S1 and a data set S2; a pure radar feature extractor M3 is trained by using a pre-trained teacher network model M2 and the data set S1, the teacher network model M2 fuses the radar data and the pulse wave data in the data set S1 and outputs fused feature data, and the pure radar feature extractor M3 performs feature extraction on the radar data in the data set S1 and outputs radar feature data; training a bimodal sleep monitoring model M4 and a pure radar modal sleep monitoring model M5 by using the data set S2, and determining first sleep stage and / or respiratory event information by a first recognition layer according to fusion feature data output by the teacher network model M2, and the second identification layer determines second sleep stage and / or respiratory event information according to the radar feature data output by the pure radar feature extractor M3. According to the invention, the sleep monitoring task is efficiently completed at low cost.

The invention belongs to the technical field of radar signal processing, and discloses a sea surface small target detection method based on optimized characteristic mode decomposition, which comprises the following steps: S1, acquiring to-be-detected signal data; s2, decomposing an original signal into a plurality of modal components by using FMD, and selecting an envelope spectrum entropy as a fitness function; s3, performing global optimization on the fitness function in the FMD by using an SOS algorithm; s4, introducing a PSO algorithm to carry out local optimization on key parameters of the FMD; s5, components with low envelope spectrum entropy values and correlation coefficients larger than a threshold value are reserved; s6, extracting an envelope spectrum entropy and frequency band energy ratio feature from the screened modal components, introducing a Gini coefficient as a weighting factor, and constructing a GSEBE joint feature; and S7, inputting the entropy value of the envelope spectrum into a DELM classifier with a controllable false alarm, and realizing target detection based on comparison between a predicted value and a judgment threshold. According to the invention, the capability of distinguishing sea clutters and target echoes is enhanced, and more accurate classification detection is realized.

The invention provides an SAR deception jamming method and device based on distance and azimuth two-dimensional modulation, and relates to the technical field of radar signal processing. The method comprises the following steps: determining an SAR signal intercepted by a jammer according to a transmitting signal of the SAR and an instantaneous slope distance from the jammer to the SAR; an interference signal is determined according to the instantaneous slant range difference between the jammer and the target, the distance offset in the distance direction, the distance offset in the azimuth direction and the SAR signal intercepted by the jammer, and the interference signal comprises an active cancellation signal and a distance and azimuth two-dimensional modulation phase; carrier frequency removal processing is carried out on the echo signal received by the SAR, the echo signal after carrier frequency removal is obtained, and the echo signal received by the SAR is the sum of the interference signal and the target echo signal; and imaging the echo signal after carrier frequency removal by using a distance Doppler algorithm to obtain an echo signal after imaging processing, so that diversified interference effects can be achieved.

The present disclosure discloses a method for suppressing the azimuth ambiguity of multi-channel SAR systems based on channel cancellation, which falls within the field of radar signal processing. The method includes: imaging echoed of the multi-channel SAR system based on a linear mapping reconstruction algorithm to obtain a reconstructed high-resolution wide-swath image; estimating positions of the azimuth ambiguity regions of the reconstructed high-resolution wide-swath image; extracting ambiguity images in the azimuth ambiguity regions based on channel cancellation and the refocusing algorithm; and suppressing a repetitive ambiguity image in imaging the echo of the multi-channel SAR system. The method weakens the influence of the original image on azimuth ambiguity extraction, improves the performance of ambiguity suppression, and solves the problem of additional repetitive ambiguity in multi-channel SAR systems by reconstructing the ambiguity signal, and improves the image quality of the multi-channel SAR system, which has great value for subsequent applications such as SAR image interpretation.

The invention discloses a laser radar echo signal distance inversion method based on a deep learning combination model, and belongs to the technical field of laser radar signal processing. According to the method, firstly, a CNN, ResNet or DCNN architecture is adopted to construct a spatial feature extraction module, and local and global spatial features of echo signals are extracted through multi-scale convolution, residual connection and an adaptive denoising mechanism; then, a time sequence dependency relationship of the signals is captured by utilizing BiGRU bidirectional time sequence modeling and a multi-head self-attention mechanism of Transform; and integrating the spatio-temporal features through the feature fusion layer, and then outputting a distance inversion value through the full connection layer. A parallel multi-scale CNN, a shrinkage enhanced residual network and a lightweight DCNN structure are innovatively designed, and the problems of insufficient spatial-temporal feature fusion of a single model and precision degradation under low signal-to-noise ratio interference are solved in combination with dynamic gating adjustment of a bidirectional gating circulation unit and dynamic position coding of Transform. According to the method, the precision and robustness of distance inversion in a complex environment are remarkably improved, and the method can be applied to the fields of automatic driving, environment monitoring and the like.

The invention is suitable for the technical field of radar signal processing, and provides a space target SFM three-dimensional reconstruction method based on a multi-frame ISAR image, and the method comprises the steps: carrying out the preprocessing and imaging of large-angle inverse synthetic aperture radar ISAR echo data of a space target, and obtaining a multi-frame ISAR image sequence; extracting scattering points in each frame of ISAR image sequence according to an orthogonal matching pursuit algorithm, and performing feature matching on adjacent frames of ISAR image sequences to obtain a scattering point set; obtaining instantaneous radar sight line information, and constructing a projection matrix from a space target scattering center three-dimensional coordinate to an ISAR two-dimensional pixel plane according to the instantaneous radar sight line information; and obtaining a three-dimensional reconstruction result according to the scattering point set and the projection matrix in combination with a triangulation principle in an incremental motion recovery structure SFM method. According to the method, the robustness and precision of three-dimensional reconstruction can be improved.

The invention relates to the technical field of computer vision and radar perception, in particular to a ship intelligent detection system based on a convolutional neural network and radar signal processing, which comprises a ship three-dimensional perception modeling module, a visual feature hierarchical fusion module, a target detection module and an anomaly detection module. The system emphatically utilizes deep learning methods such as a convolutional neural network and the like to realize three-dimensional space modeling and visual feature extraction of a ship by a radar in a water area environment. And a layered adaptive fusion and mutual information enhancement mechanism is adopted. The end-to-end detection model introduces a spatial hierarchy weighting strategy, so that the object detection accuracy and interpretability under the conditions of multi-target density, shielding and dynamic change in a complex water scene are improved. The system realizes continuous tracking, anomaly detection and risk early warning of ship navigation behaviors based on space-time dynamic modeling. The whole scheme has high precision, strong robustness and adaptive ability, and can meet the requirements of ship detection and intelligent management and control in complex water area environments such as smart ports and water traffic.

The invention discloses a human body posture estimation method based on radar point cloud imaging and multi-dimensional feature fusion, and relates to the cross technical field of computer vision and radar signal processing, and the method comprises the following three key technical links: firstly, improving the target resolution through spatial energy distribution estimation; reconstructing target three-dimensional space distribution by using the positive correlation between radar signal energy and a target reflection area and adopting a least square estimation algorithm; secondly, constructing a structured multi-dimensional point cloud matrix, and converting sparse radar point cloud into high-information-density imaging representation through a distance-speed hierarchical sorting strategy; and finally, designing a multi-dimensional feature fusion attitude estimation network, integrating three-dimensional convolution, a multi-head attention mechanism and a gating circulation unit, and realizing collaborative extraction of spatio-temporal features. According to the method, the problems of sparse target features, noise sensitivity and poor universality in traditional millimeter wave radar attitude estimation are solved.

The invention provides a strong real-time radar signal processing system, and the system comprises the steps: receiving strong real-time beam data and radar working parameters through a front-end processing module, processing the strong real-time beam data to obtain first beam data, and storing the first beam data and the radar working parameters in a CPU blade; the CPU blade receives the first beam data and the radar working parameters, and sends the first beam data and the radar working parameters to the GPU blade; the GPU blade receives the first beam data and the radar working parameters, processes the first beam data and the radar working parameters to obtain multiple groups of beam detection data, and sends the multiple groups of beam detection data to the CPU blade; the CPU blade receives and processes multiple groups of beam detection data to obtain a trace point report, and sends the trace point report to the target equipment, compared with a traditional FPGA + DSP processing architecture, the method has the advantages of being fast in development process, high in program transportability, low in cost and the like.

The invention discloses a multipath interference suppression method and device during in-pipe detection and a storage medium, and belongs to the technical field of radar signal processing. According to the method, a CFAR detection and clustering algorithm combined method is adopted, and the target number and the center frequency are automatically extracted from a radar amplitude-frequency signal; an original noisy signal is decomposed into a target number of modal components by using an improved variational modal model, and an improved power reference analysis method is proposed to divide the modal components into effective modals and ineffective modals; sVD noise reduction is carried out on the effective mode; and performing signal reconstruction on the plurality of modal components after SVD noise reduction to obtain a signal after multipath interference suppression. According to the method, the problem that the decomposition number and the center frequency of the variational mode decomposition algorithm are difficult to determine is effectively solved, the influence of inaccurate parameters is reduced, multipath interference suppression during complex environment detection is facilitated, and the effectiveness of the algorithm is verified through actual measurement.

The invention relates to the technical field of radar signal processing and quantum computing, in particular to a vehicle-mounted radar signal interference noise reduction method based on a quantum convolution region proposal network. According to the technical scheme, the vehicle-mounted radar signal interference noise reduction method based on the quantum convolution region proposal network can efficiently extract features in a vehicle-mounted radar signal by combining the parallel processing advantages of the quantum convolution region proposal network and quantum calculation, accurately identify and suppress various complex interference signals, and improve the noise reduction performance of the vehicle-mounted radar signal. Self-adaptive processing can be carried out according to the characteristics of different interference signals, it is ensured that the system can cope with variable interference sources, the parallel processing capacity of quantum computing is utilized, rapid and accurate feature extraction and interference suppression can be achieved in the signal processing process, and compared with a traditional high-cost laser radar or visual image recognition system, the system has the advantage that the system is more accurate in feature extraction and interference suppression. On the basis of the radar signal processing method and the advantage of low power consumption of quantum calculation, lower hardware cost and energy consumption can be realized.

The invention discloses a radar adaptive multi-target tracking method under dense clutters, which relates to the technical field of radar signal processing, adaptive filtering and multi-target tracking, and effectively reduces the calculation burden by simplifying a joint probability data association algorithm, reconstructing a confirmation matrix and directly using the confirmation matrix to calculate the association probability. Secondly, a self-adaptive extended Kalman filtering algorithm is introduced, an error covariance matrix is adjusted through a self-adaptive factor, and the tracking stability is improved; and finally, introducing a life cycle to perform target track management, dynamically determining the starting and ending of the track, and accurately tracking the high maneuvering target in the dense clutter environment. The problems of high calculation complexity, poor anti-interference performance and insufficient adaptability to a high-speed maneuvering target can be solved.

The invention discloses a negative group delay system based on SSPP, and belongs to the technical field of microwave communication. A fundamental mode negative group velocity is innovatively generated based on an artificial surface plasmon SSPP, and an SSPP system with negative group delay is realized. The system comprises two parts, namely an amplifier circuit and an SSPP waveguide with a negative group delay unit. The SSPP waveguide with the negative group delay characteristic comprises two coplanar waveguide parts, two transition parts and an SSPP waveguide part, the SSPP waveguide part comprises a unit for generating negative group velocity and a plurality of basic units, and further comprises a microwave dielectric substrate. The amplifier is composed of an amplifier chip, a peripheral capacitor inductor of the amplifier chip and a direct-current power supply. An input signal passes through the system, and advanced transmission of the signal can be realized. The problems of narrow bandwidth, large loss and non-adjustable structure in the prior art are effectively solved, and the antenna has wide application prospects in the fields of 5G communication, radar signal processing, array antenna systems and the like.

The invention provides a radiation source individual identification method and system based on unintentional phase modulation characteristics, and relates to the technical field of radar signal processing. The method comprises the following steps: acquiring intermediate frequency data of radar signals emitted by different radar radiation source individuals to obtain a plurality of intermediate frequency data, and extracting a plurality of monopulse signals corresponding to the plurality of intermediate frequency data from the plurality of intermediate frequency data; phase extraction is carried out on the multiple monopulse signals, and unintentional modulation phase feature maps corresponding to the multiple pieces of intermediate frequency data are obtained; and constructing a radar individual identification model and training by using the data set, and identifying the plurality of unintentional modulation phase feature maps through the trained radar individual identification model to obtain a plurality of radar radiation source individual identification results. By extracting unintentional modulation phase features of radar signals, original fingerprint information of the signals is reserved to reduce calculation complexity, and a deep residual network of a small convolution kernel is constructed as a radar individual recognition model to improve recognition accuracy.

The invention discloses a radar signal noise reduction method for deep learning, and relates to the technical field of radar signal processing, and the method comprises the steps: converting a time domain radar signal into a two-dimensional time-frequency graph through short-time Fourier transform; frequency domain feature extraction is carried out on the two-dimensional time-frequency graph, and frequency domain feature representation is output through a DnCNN and a cascade CNN in sequence; performing time domain feature extraction on the time domain radar signal and outputting time domain feature representation; fusing the frequency domain features and the time domain features based on a multi-head attention mechanism to generate cross-domain joint features; and inputting the cross-domain joint feature into a residual shrinkage network for signal reconstruction, and outputting a denoised time domain signal. A time-frequency double-domain collaborative learning framework is constructed, and signal high-fidelity reconstruction in a complex noise environment is realized through deep fusion of time domain characteristics and a frequency domain structure distribution rule. According to the noise reduction method, an extremely low phase error and an extremely high feature retention rate can be kept in a strong noise environment.

The invention is suitable for the field of radar signal processing, and provides a multifunctional radar working mode identification method based on a knowledge graph ROTATE model, and the method comprises the steps: randomly generating a radar sample data set according to the preset parameter range and parameter modulation mode of each working mode of a radar; preprocessing the radar sample data set to form a basic unit triple for constructing the knowledge graph; constructing a knowledge graph for describing a relation between a radar working mode and a parameter value and a parameter modulation type based on the basic unit triad; entities and relationships in the knowledge graph are mapped to a low-dimensional continuous vector space for representation, and the confidence coefficient of each basic unit triple is calculated by using a score function defined by an ROTATE model algorithm; and maximizing the confidence coefficient of the basic unit triad to train the model, and performing working mode recognition on the test set by using the trained model. According to the method, the recognition time can be shortened while the high recognition rate is guaranteed, and good generalization is achieved.

The invention discloses a radar target classification method based on a double-flow space-time cross attention map convolutional network, and relates to the technical field of radar signal processing and artificial intelligence cross, and the method comprises the steps: processing an obtained radar echo signal sequence, and obtaining one or more data frame sequences; for any data frame sequence, executing a first operation, specifically, taking each data frame in the target data frame sequence as a graph node, constructing a graph structure model, and determining an adjacent matrix corresponding to the graph structure model; taking the target data frame sequence and the adjacent matrix as input, and determining the probability of each target category monitored by the radar corresponding to the target data frame sequence by using the trained double-flow space-time cross attention map convolutional network; and determining the target category corresponding to the maximum probability as a radar monitoring target category corresponding to the target data frame sequence. According to the invention, the accuracy and anti-interference performance of radar target classification can be improved.

The invention discloses a non-contact electrocardiosignal monitoring reconstruction method and system based on time-frequency domain combination, and relates to the technical field of radar signal processing, and the method comprises the following steps: obtaining a chest wall vibration signal S of a to-be-detected person by adopting a frequency-modulated continuous wave radar; separating a heart mechanical motion signal SH from the chest wall vibration signal S; and converting the heart mechanical motion signal SH into an electrocardiosignal ECG through a pre-established ResNet architecture neural network model. Time domain loss and frequency domain loss are used for training a ResNet architecture neural network model, and the error between the calculated heart rate and the real heart rate is quantified through the mean absolute error and the root-mean-square error. According to the method, the neural network capable of being used for ECG signal conversion is designed based on the ResNet architecture, after the neural network is subjected to data training, more detailed signal features can be learned, the fitting accuracy is improved, then the ECG waveform can be accurately predicted, and non-contact electrocardiosignal monitoring and reconstruction are achieved.

The invention belongs to the technical field of intelligent wireless sensing and radar signal processing, and particularly relates to a millimeter wave radar gesture recognition feature processing method, which is particularly suitable for scenes with environment interference (such as walking of others and static clutter), and specifically comprises the following steps: S1, preliminary filtering by a self-adaptive filter; s2, carrying out improved I CEEMDAN decomposition; s3, I MF component classification and processing; s4, signal reconstruction; the experimental result shows that the method provides a robust and efficient solution for gesture recognition, and can be widely applied to the fields of man-machine interaction, virtual reality, intelligent equipment control and the like.

The invention discloses an airborne radar system and method oriented to a signal transmitting device, and belongs to the technical field of airborne radar signal processing. Received radar task instructions are analyzed, target characteristic parameters, environment constraint conditions and performance indexes are extracted, and mapping relations between task requirements and transmitting waveforms, power and time sequence parameters are established; based on the mapping relation, the transmitting device resources are modeled into a power, frequency spectrum and time sequence three-dimensional state space, and the resource margin and the conflict risk are monitored and predicted in real time; matching a current task situation with a historical strategy library in combination with a resource state, and selecting a historical strategy with the highest similarity as a basic launching scheme; optimizing a resource allocation demand of the basic scheme according to a time sequence and a priority sequence, and introducing a resource buffer mechanism and dynamic priority adjustment to solve a multi-task resource conflict; based on real-time environment data, a basic scheme is decomposed into independently adjustable strategy units, and after optimization and recombination, a transmitting strategy adaptive to a specific scene is generated.

The invention discloses a non-uniform clutter and interference cooperative suppression target fusion detection method and system, and relates to the technical field of broadband radar signal processing, and the method comprises the steps: carrying out the unitary transformation of test data, a clutter covariance matrix, a target coordinate matrix and an interference coordinate matrix based on the skew symmetry of the clutter covariance matrix; constructing a distance extension target Gradient detection statistical magnitude under the condition of a known clutter skew symmetry covariance matrix; calculating the maximum likelihood estimation of the clutter skew symmetry covariance matrix; based on the maximum likelihood estimation of the clutter skew-symmetric covariance matrix and the distance expansion target Gradant detection statistic, constructing a target detection statistic of non-uniform clutter and interference cooperative suppression; performing target fusion detection on the target detection unit based on the target detection statistical magnitude and a preset detection threshold; according to the invention, the technical problems of complex construction process and high calculation complexity of the target detector in the prior art are solved.

The invention discloses a satellite-borne inverse synthetic aperture radar motion compensation imaging calibration integration method, equipment and a medium, and relates to the technical field of radar signal processing and radar imaging. The objective of the invention is to solve the problems of inaccurate motion compensation and difficult large-rotation-angle imaging in ISAR imaging. According to the method, a particle swarm optimization algorithm is adopted to optimize motion parameters, optimal motion parameters are obtained, and the motion parameters comprise relative translation motion parameters and relative rotation motion parameters of a target and a radar; and obtaining focusing and calibration imaging results based on the optimal motion parameters and echo data reflected by the target. The influence of the translation compensation effect on rotation parameter estimation is reduced, more accurate parameter estimation and better imaging focusing performance are realized, and the ISAR imaging problem under the condition of violent relative movement in the rendezvous scene is solved.

The invention relates to a sea surface target identification method applying time domain approximate entropy features, and belongs to the technical field of radar signal processing and feature identification. Comprising the steps of 1, caching target echo data; 2, performing time domain analysis on echo data extracted from target time domain approximate entropy features; step 3, time domain approximate entropy feature optimization; and 4, target identification is carried out by using the time domain approximate entropy features. The objective of the invention is to identify a sea surface target under the condition that target identification is difficult to effectively perform through traditional target features.

The invention discloses a semi-supervised LPI radar signal modulation identification system and method based on entropy perception pseudo labels, and relates to the technical field of radar signal processing and mode identification. The system comprises a preprocessing module, a multi-scale reconstruction enhancer, a classification backbone network and a semi-supervised training module. The multi-scale reconstruction intensifier is used for reconstructing dual-channel separation through high-frequency detail enhancement and a low-frequency structure and enhancing discriminative characteristics in a noise environment; the classification backbone network introduces an adaptive contraction unit to realize channel-level noise suppression; and the semi-supervised training module dynamically evaluates the uncertainty of the unlabeled samples by adopting an entropy sensing mechanism, and generates weighted pseudo labels to carry out consistency regularization training. The method realizes signal modulation identification based on the system. According to the method, the problem of feature shielding under the condition of low signal-to-noise ratio is solved, the dependence of the model on labeled data is reduced through a reliable pseudo label generation mechanism, and stable and efficient modulation identification can still be realized in a severe channel environment with scarce labeled data.