Submit Manuscript
Peer Review
Editor WorkAccepted Article Preview
Abstract
20668KB
To address issues such as insufficient feature extraction, limited spatiotemporal correlation modeling, and poor classification performance in radar classification of low, slow, and small targets, this paper investigates on graph network-based feature extraction and classification methods. First, focusing on digital array ubiquitous radar, a radar detection dataset for LSS targets, named LSS-DAUR-1.0, is constructed; it contains Doppler and track data for six types of targets: passenger ships, speedboats, helicopters, rotor drones, birds, and fixed-wing drones. Second, based on this dataset, the multidomain and multidimensional characteristics of the targets are analyzed, and the complementarity between Doppler and physical motion features is verified through correlation and cosine similarity analyses. On this basis, a Graph Convolutional Network with Dynamic Graph Construction (DG-GCN) classification method fusing dual features is proposed. An adaptive window adjustment, a hybrid attenuation function, and a dynamic threshold mechanism are designed to construct an adaptive dynamic graph based on spatiotemporal correlation. Combined with graph convolution–based feature learning and classification modules, this approach achieves refined classification of low, slow, and small targets. Validation on the LSS-DAUR-1.0 dataset shows that the DG-GCN achieves 99.66% classification accuracy, which is 6.78% and 17.97% higher than that of ResNet and Transformer models, respectively. The total processing time is only 4.98 ms, which is more than 80% lower than that of the aforementioned comparison models. Hence, the DG-GCN achieves both high accuracy and efficiency. In addition, noise environment tests show good robustness. Ablation experiments verify that the dynamic edge weight mechanism compensates for the lack of spatial feature correlation in purely temporal connections and improves the model’s generalizability.
To address issues such as insufficient feature extraction, limited spatiotemporal correlation modeling, and poor classification performance in radar classification of low, slow, and small targets, this paper investigates on graph network-based feature extraction and classification methods. First, focusing on digital array ubiquitous radar, a radar detection dataset for LSS targets, named LSS-DAUR-1.0, is constructed; it contains Doppler and track data for six types of targets: passenger ships, speedboats, helicopters, rotor drones, birds, and fixed-wing drones. Second, based on this dataset, the multidomain and multidimensional characteristics of the targets are analyzed, and the complementarity between Doppler and physical motion features is verified through correlation and cosine similarity analyses. On this basis, a Graph Convolutional Network with Dynamic Graph Construction (DG-GCN) classification method fusing dual features is proposed. An adaptive window adjustment, a hybrid attenuation function, and a dynamic threshold mechanism are designed to construct an adaptive dynamic graph based on spatiotemporal correlation. Combined with graph convolution–based feature learning and classification modules, this approach achieves refined classification of low, slow, and small targets. Validation on the LSS-DAUR-1.0 dataset shows that the DG-GCN achieves 99.66% classification accuracy, which is 6.78% and 17.97% higher than that of ResNet and Transformer models, respectively. The total processing time is only 4.98 ms, which is more than 80% lower than that of the aforementioned comparison models. Hence, the DG-GCN achieves both high accuracy and efficiency. In addition, noise environment tests show good robustness. Ablation experiments verify that the dynamic edge weight mechanism compensates for the lack of spatial feature correlation in purely temporal connections and improves the model’s generalizability.
With the widespread use of millimeter-wave radar technology in indoor target detection and tracking, multipath effects have become a key factor affecting tracking accuracy. Indoor millimeter-wave radar target tracking is highly susceptible to multipath interference, and conventional point-target tracking methods, which ignore the extended characteristics of targets and the multipath propagation mechanism, struggle to effectively suppress ghost targets caused by multipath reflections. To address this issue, this paper proposes an extension mapping-based extended target tracking (EM-ETT) method for indoor target tracking using millimeter-wave radar. First, a random matrix model is used to characterize the target’s geometric shape, with the extension modeled as an inverse Wishart distribution. Next, an extended projection framework is constructed by integrating a Monte Carlo-based statistical propagation mechanism. Through nonlinear multipath mapping of scattering points from the true target, ghost point clouds are generated, and their extended state priors are estimated. Furthermore, a target–path association method is introduced to establish path associations in multipath propagation based on geometric consistency and likelihood evaluation, enhancing state discrimination capability. Experimental results demonstrate that in multitarget scenarios with multipath interference, the proposed method significantly improves state estimation accuracy and effectively prevents the generation of false trajectories. Compared with conventional point-target tracking algorithms, the proposed method exhibits significant advantages in both tracking accuracy and robustness.
With the widespread use of millimeter-wave radar technology in indoor target detection and tracking, multipath effects have become a key factor affecting tracking accuracy. Indoor millimeter-wave radar target tracking is highly susceptible to multipath interference, and conventional point-target tracking methods, which ignore the extended characteristics of targets and the multipath propagation mechanism, struggle to effectively suppress ghost targets caused by multipath reflections. To address this issue, this paper proposes an extension mapping-based extended target tracking (EM-ETT) method for indoor target tracking using millimeter-wave radar. First, a random matrix model is used to characterize the target’s geometric shape, with the extension modeled as an inverse Wishart distribution. Next, an extended projection framework is constructed by integrating a Monte Carlo-based statistical propagation mechanism. Through nonlinear multipath mapping of scattering points from the true target, ghost point clouds are generated, and their extended state priors are estimated. Furthermore, a target–path association method is introduced to establish path associations in multipath propagation based on geometric consistency and likelihood evaluation, enhancing state discrimination capability. Experimental results demonstrate that in multitarget scenarios with multipath interference, the proposed method significantly improves state estimation accuracy and effectively prevents the generation of false trajectories. Compared with conventional point-target tracking algorithms, the proposed method exhibits significant advantages in both tracking accuracy and robustness.
Vortex electromagnetic (EM) wave radars utilize EM waves carrying orbital angular momentum to enrich target scattering information, thereby providing intrinsic in-beam azimuth resolution. Hence, this technology holds significant potential for advanced target detection and imaging. However, as sensing scenarios become more complex, conventional electronic vortex EM wave radars are increasingly limited by device bandwidth. Specifically, they encounter substantial challenges in broadband signal generation and control, making it difficult to achieve high range and azimuth resolutions simultaneously. Microwave photonics technology, with its inherent advantages of wide bandwidth, low transmission loss, and robustness against electromagnetic interference, is an effectivesolution to overcome these limitations. This paper reviews recent progress in microwave photonic broadband vortex EM wave radars, addressing the requirements for forward-looking imaging. The fundamental system architectures and imaging mechanisms are elucidated, followed by a critical analysis of the frequency-dependent characteristics of broadband vortex waves and their implications for imaging performance. Key microwave photonic enabling technologies, including broadband phase shifting, optical beamforming, and broadband signal generation, are summarized, and their advantages over traditional electronic schemes in terms of performance are highlighted. Based on these insights, typical system implementation schemes are described, and their high-resolution forward-looking imaging capabilities are demonstrated through proof-of-concept experiments. Finally, future development trends and research directions are discussed.
Vortex electromagnetic (EM) wave radars utilize EM waves carrying orbital angular momentum to enrich target scattering information, thereby providing intrinsic in-beam azimuth resolution. Hence, this technology holds significant potential for advanced target detection and imaging. However, as sensing scenarios become more complex, conventional electronic vortex EM wave radars are increasingly limited by device bandwidth. Specifically, they encounter substantial challenges in broadband signal generation and control, making it difficult to achieve high range and azimuth resolutions simultaneously. Microwave photonics technology, with its inherent advantages of wide bandwidth, low transmission loss, and robustness against electromagnetic interference, is an effectivesolution to overcome these limitations. This paper reviews recent progress in microwave photonic broadband vortex EM wave radars, addressing the requirements for forward-looking imaging. The fundamental system architectures and imaging mechanisms are elucidated, followed by a critical analysis of the frequency-dependent characteristics of broadband vortex waves and their implications for imaging performance. Key microwave photonic enabling technologies, including broadband phase shifting, optical beamforming, and broadband signal generation, are summarized, and their advantages over traditional electronic schemes in terms of performance are highlighted. Based on these insights, typical system implementation schemes are described, and their high-resolution forward-looking imaging capabilities are demonstrated through proof-of-concept experiments. Finally, future development trends and research directions are discussed.
Heart Rate (HR), a core physiological indicator of human health, is of substantial clinical importance when accurately monitored in applications such as arrhythmia screening, early warning of coronary heart disease, and chronic heart failure management. However, cardiac echo signals are susceptible to coupled disturbances, including respiratory motion artifacts and environmental electromagnetic interference, which degrade the signal-to-noise ratio and compromise HR estimation accuracy. To address these challenges, we propose a multi-channel joint HR estimation method based on multivariate variational mode decomposition that exploits shared cardiac information across different channels. Specifically, the proposed method first constructs a multi-channel joint optimization model that minimizes the total modal bandwidth under reconstruction residual constraints. It then adaptively initialize the center frequency by leveraging the cumulative effect of multi-channel spectral peaks, enabling robust separation of heartbeat modes with consistent frequencies across channels. Finally, the HR mode is selected from the decomposed modes using a maximum energy criterion to complete HR estimation. Validation on real-world data from six subjects demonstrated that the proposed method achieves a median HR error of 1.53 bpm, outperforming conventional single-channel approaches and existing multi-channel fusion-based HR estimation methods.
Heart Rate (HR), a core physiological indicator of human health, is of substantial clinical importance when accurately monitored in applications such as arrhythmia screening, early warning of coronary heart disease, and chronic heart failure management. However, cardiac echo signals are susceptible to coupled disturbances, including respiratory motion artifacts and environmental electromagnetic interference, which degrade the signal-to-noise ratio and compromise HR estimation accuracy. To address these challenges, we propose a multi-channel joint HR estimation method based on multivariate variational mode decomposition that exploits shared cardiac information across different channels. Specifically, the proposed method first constructs a multi-channel joint optimization model that minimizes the total modal bandwidth under reconstruction residual constraints. It then adaptively initialize the center frequency by leveraging the cumulative effect of multi-channel spectral peaks, enabling robust separation of heartbeat modes with consistent frequencies across channels. Finally, the HR mode is selected from the decomposed modes using a maximum energy criterion to complete HR estimation. Validation on real-world data from six subjects demonstrated that the proposed method achieves a median HR error of 1.53 bpm, outperforming conventional single-channel approaches and existing multi-channel fusion-based HR estimation methods.
Sea surface elevation is crucial for characterizing individual waves, wave groups, and freak waves, offering an accurate representation of inhomogeneous sea states. This study presents a quasi-linear inversion strategy for retrieving sea surface elevation from GF-3 Synthetic Aperture Radar (SAR) images. The algorithm enables rapid inversion within 10 s per scene without the need for external data and effectively resolves range-traveling waves. Case studies conducted under three distinct sea states demonstrate its ability to extract maximum wave heights and identify wave groups. Additionally, inversion results from 2405 GF-3 wave mode SAR images following quality control are compared with ERA5 reanalysis spectra and altimeter data. The comparisons reveal that the retrieved Significant Wave Height (SWH) has a root mean square error of 0.48 m compared with ERA5 data. In low-to-moderate sea states, with significant wave heights below 3 m, the retrieved SWH shows strong consistency with ERA5 spectra and altimeter measurements. This algorithm serves as an effective tool for rapid monitoring and analysis of sea states using GF-3 SAR.
Sea surface elevation is crucial for characterizing individual waves, wave groups, and freak waves, offering an accurate representation of inhomogeneous sea states. This study presents a quasi-linear inversion strategy for retrieving sea surface elevation from GF-3 Synthetic Aperture Radar (SAR) images. The algorithm enables rapid inversion within 10 s per scene without the need for external data and effectively resolves range-traveling waves. Case studies conducted under three distinct sea states demonstrate its ability to extract maximum wave heights and identify wave groups. Additionally, inversion results from 2405 GF-3 wave mode SAR images following quality control are compared with ERA5 reanalysis spectra and altimeter data. The comparisons reveal that the retrieved Significant Wave Height (SWH) has a root mean square error of 0.48 m compared with ERA5 data. In low-to-moderate sea states, with significant wave heights below 3 m, the retrieved SWH shows strong consistency with ERA5 spectra and altimeter measurements. This algorithm serves as an effective tool for rapid monitoring and analysis of sea states using GF-3 SAR.
Synthetic Aperture Radar (SAR) is widely used in military and civilian applications, with intelligent target interpretation of SAR images being a crucial component of SAR applications. Vision-language Models (VLMs) play an important role in SAR target interpretation. By incorporating natural language understanding, VLMs effectively address the challenges posed by large intraclass variability in target characteristics and the scarcity of high-quality labeled samples, thereby advancing the field from purely visual interpretation toward semantic understanding of targets. Drawing upon our team’s extensive research experience in SAR target interpretation theory, algorithms, and applications, this paper provides a comprehensive review of intelligent SAR target interpretation based on VLMs. We provide an in-depth analysis of existing challenges and tasks, summarize the current state of research, and compile available open-source datasets. Furthermore, we systematically outline the evolution, ranging from task-specific VLMs to contrastive-, conversational-, and generative-based VLMs and foundational models. Finally, we discuss the latest challenges and future outlooks in SAR target interpretation by VLMs.
Synthetic Aperture Radar (SAR) is widely used in military and civilian applications, with intelligent target interpretation of SAR images being a crucial component of SAR applications. Vision-language Models (VLMs) play an important role in SAR target interpretation. By incorporating natural language understanding, VLMs effectively address the challenges posed by large intraclass variability in target characteristics and the scarcity of high-quality labeled samples, thereby advancing the field from purely visual interpretation toward semantic understanding of targets. Drawing upon our team’s extensive research experience in SAR target interpretation theory, algorithms, and applications, this paper provides a comprehensive review of intelligent SAR target interpretation based on VLMs. We provide an in-depth analysis of existing challenges and tasks, summarize the current state of research, and compile available open-source datasets. Furthermore, we systematically outline the evolution, ranging from task-specific VLMs to contrastive-, conversational-, and generative-based VLMs and foundational models. Finally, we discuss the latest challenges and future outlooks in SAR target interpretation by VLMs.
Research on target recognition using radar High-Resolution Range Profiles (HRRPs) is extensive and diverse in methodology. In particular, the application and development of deep learning to radar HRRP target recognition have enabled efficient, precise target perception directly from radar echoes. However, deep learning-based recognition networks rely on large amounts of training data. For non-cooperative targets, due to limited radar system parameters and rapid target attitude variations, acquiring adequate HRRP training samples that comprehensively cover target attitudes in advance is difficult in practice. Consequently, deep recognition networks are prone to overfitting and exhibit considerably degraded generalization capability. To address these issues, and given the ease of obtaining full-attitude electromagnetic simulation data for the target, this paper leverages simulated data as auxiliary information to mitigate the small-sample-size problem through data augmentation and cross-domain knowledge-transfer learning. For data augmentation, based on the analysis of differences in mean and variance between simulated and measured HRRPs within a given attitude-angle range, a linear transformation is applied to a set of simulated HRRPs spanning the same angular domain as a small set of measured HRRPs. This adjustment ensures that the simulated data’s mean and variance match the characteristics of the measured HRRPs, thereby achieving data augmentation that approximates the true distributional properties of HRRPs. Meanwhile, for cross-domain knowledge transfer learning, the proposed method introduces a domain alignment strategy based on generative adversarial constraints and a class alignment strategy based on contrastive learning constraints. These approaches draw the domain features of full-attitude simulation—strong discriminability and generalizability—closer to the measured domain features on a class-by-class basis, thereby further aiding learning from the measured domain data and leading to substantial improvements in few-shot recognition performance. Experimental results based on electromagnetic simulated and measured HRRP data for three and ten types of aircraft and ground vehicle targets, respectively, demonstrate that the proposed method yields superior recognition robustness compared with existing few-shot recognition methods.
Research on target recognition using radar High-Resolution Range Profiles (HRRPs) is extensive and diverse in methodology. In particular, the application and development of deep learning to radar HRRP target recognition have enabled efficient, precise target perception directly from radar echoes. However, deep learning-based recognition networks rely on large amounts of training data. For non-cooperative targets, due to limited radar system parameters and rapid target attitude variations, acquiring adequate HRRP training samples that comprehensively cover target attitudes in advance is difficult in practice. Consequently, deep recognition networks are prone to overfitting and exhibit considerably degraded generalization capability. To address these issues, and given the ease of obtaining full-attitude electromagnetic simulation data for the target, this paper leverages simulated data as auxiliary information to mitigate the small-sample-size problem through data augmentation and cross-domain knowledge-transfer learning. For data augmentation, based on the analysis of differences in mean and variance between simulated and measured HRRPs within a given attitude-angle range, a linear transformation is applied to a set of simulated HRRPs spanning the same angular domain as a small set of measured HRRPs. This adjustment ensures that the simulated data’s mean and variance match the characteristics of the measured HRRPs, thereby achieving data augmentation that approximates the true distributional properties of HRRPs. Meanwhile, for cross-domain knowledge transfer learning, the proposed method introduces a domain alignment strategy based on generative adversarial constraints and a class alignment strategy based on contrastive learning constraints. These approaches draw the domain features of full-attitude simulation—strong discriminability and generalizability—closer to the measured domain features on a class-by-class basis, thereby further aiding learning from the measured domain data and leading to substantial improvements in few-shot recognition performance. Experimental results based on electromagnetic simulated and measured HRRP data for three and ten types of aircraft and ground vehicle targets, respectively, demonstrate that the proposed method yields superior recognition robustness compared with existing few-shot recognition methods.
Ultra-Wideband (UWB) Multiple-Input Multiple-Output (MIMO) radar has demonstrated enormous potential in the field of human intelligent perception due to its excellent resolution, strong penetration capability, strong privacy protection, and insensitivity to illumination conditions. However, its low image resolution results in blurred contours and indistinguishable actions. To address this issue, this study developes a joint framework, Spatiotemporal Wavelet Transformer network (STWTnet), for human contour restoration and action recognition by integrating spatiotemporal features. By adopting a multi-task network architecture, the proposed framework leverages Res2Net and wavelet downsampling to extract spatial detail features from radar images and employs a Transformer to establish spatiotemporal dependencies. Through multi-task learning, it shares the common features of human contour restoration and action recognition, enabling mutual complementarity between the two tasks while avoiding feature conflicts. Experiments conducted on a self-built, synchronized UWB optical dataset demonstrate that STWTnet achieves high action recognition accuracy and significantly outperforms existing techniques in contour restoration precision, providing a new approach for privacy-preserving, all-weather human behavior understanding.
Ultra-Wideband (UWB) Multiple-Input Multiple-Output (MIMO) radar has demonstrated enormous potential in the field of human intelligent perception due to its excellent resolution, strong penetration capability, strong privacy protection, and insensitivity to illumination conditions. However, its low image resolution results in blurred contours and indistinguishable actions. To address this issue, this study developes a joint framework, Spatiotemporal Wavelet Transformer network (STWTnet), for human contour restoration and action recognition by integrating spatiotemporal features. By adopting a multi-task network architecture, the proposed framework leverages Res2Net and wavelet downsampling to extract spatial detail features from radar images and employs a Transformer to establish spatiotemporal dependencies. Through multi-task learning, it shares the common features of human contour restoration and action recognition, enabling mutual complementarity between the two tasks while avoiding feature conflicts. Experiments conducted on a self-built, synchronized UWB optical dataset demonstrate that STWTnet achieves high action recognition accuracy and significantly outperforms existing techniques in contour restoration precision, providing a new approach for privacy-preserving, all-weather human behavior understanding.
In locating ground moving radiating sources, traditional passive positioning methods, such as Direction of Arrival (DOA), often rely on long-term observation and filtering, resulting in low positioning efficiency. Existing synthetic aperture–based positioning methods are primarily designed for stationary radiating sources, making high-precision positioning of moving sources difficult. To address this limitation, this paper proposes synthetic aperture–based fast positioning and velocity estimation methods for moving radiating sources under single- and dual-station positioning systems, respectively. The proposed methods establish an instantaneous slant range model of the radiating source and derive the mapping relationship between the positioning parameters (position and velocity) and the imaging parameters. Specifically, in the single-station scenario, the traditional second-order slant range model is extended to third order, and a third-order chirp rate is introduced to supplement the degrees of freedom, thereby enabling simultaneous estimation of position and velocity. In the dual-station scenario, an additional observation station is used to introduce two new imaging parameters, thereby further improving the rapidity and accuracy of positioning. To address the multi-solution problem inherent in the positioning equations, this paper proposes true-solution determination criteria for the single- and dual-station systems and presents an initialization strategy to ensure a unique solution for dual-station positioning. Furthermore, the paper analyzes how various factors affect the positioning accuracy of single- and dual-station models, compares the performance of the proposed single- and dual-station passive positioning models, and verifies the effectiveness of the proposed algorithms through simulations.
In locating ground moving radiating sources, traditional passive positioning methods, such as Direction of Arrival (DOA), often rely on long-term observation and filtering, resulting in low positioning efficiency. Existing synthetic aperture–based positioning methods are primarily designed for stationary radiating sources, making high-precision positioning of moving sources difficult. To address this limitation, this paper proposes synthetic aperture–based fast positioning and velocity estimation methods for moving radiating sources under single- and dual-station positioning systems, respectively. The proposed methods establish an instantaneous slant range model of the radiating source and derive the mapping relationship between the positioning parameters (position and velocity) and the imaging parameters. Specifically, in the single-station scenario, the traditional second-order slant range model is extended to third order, and a third-order chirp rate is introduced to supplement the degrees of freedom, thereby enabling simultaneous estimation of position and velocity. In the dual-station scenario, an additional observation station is used to introduce two new imaging parameters, thereby further improving the rapidity and accuracy of positioning. To address the multi-solution problem inherent in the positioning equations, this paper proposes true-solution determination criteria for the single- and dual-station systems and presents an initialization strategy to ensure a unique solution for dual-station positioning. Furthermore, the paper analyzes how various factors affect the positioning accuracy of single- and dual-station models, compares the performance of the proposed single- and dual-station passive positioning models, and verifies the effectiveness of the proposed algorithms through simulations.
Synthetic Aperture Radar (SAR), as an active microwave remote sensing system, offers all-weather, all-day observation capabilities and has considerable application value in disaster monitoring, urban management, and military reconnaissance. Although deep learning techniques have achieved remarkable progress in interpreting SAR images, existing methods for target recognition and detection primarily focus on local feature extraction and single-target discrimination. They struggle to comprehensively characterize the global semantic structure and multitarget relationships in complex scenes, and the interpretation process remains highly dependent on human expertise with limited automation. SAR image captioning aims to translate visual information into natural language, serving as a key technology to bridge the gap between “perceiving targets” and “cognizing scenes,” which is of great importance for enhancing the automation and intelligence of SAR image interpretation. However, the inherent speckle noise, the scarcity of textural details, and the substantial semantic gap in SAR images further exacerbate the difficulty of cross-modal understanding. To address these challenges, this paper proposes a spatial-frequency aware model for SAR image captioning. First, a spatial-frequency aware module is constructed. It employs a Discrete Cosine Transform (DCT) mask attention mechanism to reweight spectral components for noise suppression and structure enhancement, combined with a Gabor multiscale texture enhancement submodule to improve sensitivity to directional and edge details. Second, a cross-modal semantic enhancement loss function is designed to bridge the semantic gap between visual features and natural language through bidirectional image-text alignment and mutual information maximization. Furthermore, a large-scale fine-grained SAR image captioning dataset, FSAR-Cap, containing 72400 high-quality image-text pairs, is constructed. The experimental results demonstrate that the proposed method achieves CIDEr scores of 151.00 and 95.14 on the SARLANG and FSAR-Cap datasets, respectively. Qualitatively, the model effectively suppresses hallucinations and accurately captures fine-grained spatial-textural details, considerably outperforming mainstream methods.
Synthetic Aperture Radar (SAR), as an active microwave remote sensing system, offers all-weather, all-day observation capabilities and has considerable application value in disaster monitoring, urban management, and military reconnaissance. Although deep learning techniques have achieved remarkable progress in interpreting SAR images, existing methods for target recognition and detection primarily focus on local feature extraction and single-target discrimination. They struggle to comprehensively characterize the global semantic structure and multitarget relationships in complex scenes, and the interpretation process remains highly dependent on human expertise with limited automation. SAR image captioning aims to translate visual information into natural language, serving as a key technology to bridge the gap between “perceiving targets” and “cognizing scenes,” which is of great importance for enhancing the automation and intelligence of SAR image interpretation. However, the inherent speckle noise, the scarcity of textural details, and the substantial semantic gap in SAR images further exacerbate the difficulty of cross-modal understanding. To address these challenges, this paper proposes a spatial-frequency aware model for SAR image captioning. First, a spatial-frequency aware module is constructed. It employs a Discrete Cosine Transform (DCT) mask attention mechanism to reweight spectral components for noise suppression and structure enhancement, combined with a Gabor multiscale texture enhancement submodule to improve sensitivity to directional and edge details. Second, a cross-modal semantic enhancement loss function is designed to bridge the semantic gap between visual features and natural language through bidirectional image-text alignment and mutual information maximization. Furthermore, a large-scale fine-grained SAR image captioning dataset, FSAR-Cap, containing 72400 high-quality image-text pairs, is constructed. The experimental results demonstrate that the proposed method achieves CIDEr scores of 151.00 and 95.14 on the SARLANG and FSAR-Cap datasets, respectively. Qualitatively, the model effectively suppresses hallucinations and accurately captures fine-grained spatial-textural details, considerably outperforming mainstream methods.
The Moon’s shallow subsurface structure provides crucial insights into its geological evolution, material composition, and space weathering processes. With the acquisition of extensive radar datasets from recent lunar exploration programs, such as the Chang’E missions, high-resolution characterization of the lunar regolith’s stratigraphic and physical properties has become a focus and challenge in lunar science. Conventional radar layer identification and tracking methods often suffer from instability in complex scattering environments, because of their sensitivity to noise and subsurface heterogeneity. To address these limitations, this study proposes an automatic layer-tracking algorithm based on a Dynamic Search Center (DSC) approach. This algorithm employs a Gaussian-weighted prediction mechanism to balance historical trajectory trends with local signal responses and uses a multifeatured fusion decision scheme to enhance tracking robustness under noisy conditions. Numerical simulations demonstrate that, with a search radius l = 20 and a historical window n = 20, the algorithm achieves a layer identification error of less than 2% for shallow strata (<140 ns). Meanwhile, for deep layers (>170 ns), with considerable signal attenuation, incorporating an edge-direction weighting term reduces the tracking error by over 30%. When applied to lunar penetrating radar data from the Chang’E-4 mission, the proposed method successfully realizes automatic stratigraphic tracing in lunar radar profiles, producing layer boundaries that are highly consistent with previous interpretations. Simulation and in-situ results confirm that the DSC-based algorithm accurately delineates real subsurface interfaces across media and structural morphologies, effectively suppressing noise, while maintaining smooth trajectories. Overall, the proposed method achieves low manual dependence, high robustness, and high precision in automatic radar layer tracking, thereby providing a valuable reference for analyzing radar data from upcoming missions such as Chang’E-7 and Martian shallow-subsurface explorations.
The Moon’s shallow subsurface structure provides crucial insights into its geological evolution, material composition, and space weathering processes. With the acquisition of extensive radar datasets from recent lunar exploration programs, such as the Chang’E missions, high-resolution characterization of the lunar regolith’s stratigraphic and physical properties has become a focus and challenge in lunar science. Conventional radar layer identification and tracking methods often suffer from instability in complex scattering environments, because of their sensitivity to noise and subsurface heterogeneity. To address these limitations, this study proposes an automatic layer-tracking algorithm based on a Dynamic Search Center (DSC) approach. This algorithm employs a Gaussian-weighted prediction mechanism to balance historical trajectory trends with local signal responses and uses a multifeatured fusion decision scheme to enhance tracking robustness under noisy conditions. Numerical simulations demonstrate that, with a search radius l = 20 and a historical window n = 20, the algorithm achieves a layer identification error of less than 2% for shallow strata (<140 ns). Meanwhile, for deep layers (>170 ns), with considerable signal attenuation, incorporating an edge-direction weighting term reduces the tracking error by over 30%. When applied to lunar penetrating radar data from the Chang’E-4 mission, the proposed method successfully realizes automatic stratigraphic tracing in lunar radar profiles, producing layer boundaries that are highly consistent with previous interpretations. Simulation and in-situ results confirm that the DSC-based algorithm accurately delineates real subsurface interfaces across media and structural morphologies, effectively suppressing noise, while maintaining smooth trajectories. Overall, the proposed method achieves low manual dependence, high robustness, and high precision in automatic radar layer tracking, thereby providing a valuable reference for analyzing radar data from upcoming missions such as Chang’E-7 and Martian shallow-subsurface explorations.
To address the urgent need to identify birds and rotary-wing unmanned aerial vehicles (UAVs), this paper proposes a vortex radar–based method for extracting micromotion parameters of targets. The study focused on target parameter acquisition and systematically extended target modeling and parameter extraction strategies. First, mathematical models were developed for the body motion and wing flapping behavior of birds as well as for the rotor movement characteristics and body structure of rotary-wing UAVs. Further, analytical expressions for the radial and rotational Doppler frequency shifts at scattering points were derived, and micro-Doppler features were extracted from radar echo signals to enable target parameter inversion. For bird targets, the radial Doppler frequency was estimated by extracting the spectral peak of the echo signal to obtain the flight velocity. In addition, by combining the rotational Doppler frequency shifts of the scattering points and analyzing the variations of the rotational Doppler frequency using the short-time Fourier transform (STFT), the wing-flapping length was estimated. Even under low signal-to-noise ratio (SNR) conditions, the estimation error of the wing-flapping length remained within 0.03 m. For rotary-wing UAV targets, an echo signal model was first constructed, and the analytical relationship between the radial and rotational components of the micro-Doppler frequency shift was derived. Using the reconstructed Doppler information and through range–time domain analysis, six structural and motion parameters were retrieved, including the Euler angles rotor rotational speed, rotor length, and the distance between the UAV body and rotor. The estimation errors for all parameters were significantly lower than those obtained with conventional approaches based on individual Doppler features, with all parameters remaining within 2%. Simulation results demonstrated that the proposed vortex radar–based parameter extraction method enables accurate multiparameter estimation for birds and rotary-wing UAVs. The method also exhibits stable and reliable performance under low SNR conditions, confirming its effectiveness and applicability in practical engineering scenarios.
To address the urgent need to identify birds and rotary-wing unmanned aerial vehicles (UAVs), this paper proposes a vortex radar–based method for extracting micromotion parameters of targets. The study focused on target parameter acquisition and systematically extended target modeling and parameter extraction strategies. First, mathematical models were developed for the body motion and wing flapping behavior of birds as well as for the rotor movement characteristics and body structure of rotary-wing UAVs. Further, analytical expressions for the radial and rotational Doppler frequency shifts at scattering points were derived, and micro-Doppler features were extracted from radar echo signals to enable target parameter inversion. For bird targets, the radial Doppler frequency was estimated by extracting the spectral peak of the echo signal to obtain the flight velocity. In addition, by combining the rotational Doppler frequency shifts of the scattering points and analyzing the variations of the rotational Doppler frequency using the short-time Fourier transform (STFT), the wing-flapping length was estimated. Even under low signal-to-noise ratio (SNR) conditions, the estimation error of the wing-flapping length remained within 0.03 m. For rotary-wing UAV targets, an echo signal model was first constructed, and the analytical relationship between the radial and rotational components of the micro-Doppler frequency shift was derived. Using the reconstructed Doppler information and through range–time domain analysis, six structural and motion parameters were retrieved, including the Euler angles rotor rotational speed, rotor length, and the distance between the UAV body and rotor. The estimation errors for all parameters were significantly lower than those obtained with conventional approaches based on individual Doppler features, with all parameters remaining within 2%. Simulation results demonstrated that the proposed vortex radar–based parameter extraction method enables accurate multiparameter estimation for birds and rotary-wing UAVs. The method also exhibits stable and reliable performance under low SNR conditions, confirming its effectiveness and applicability in practical engineering scenarios.
A maritime multimodal data resource system provides a foundation for multisensor collaborative detection using radar, Synthetic Aperture Radar (SAR), and electro-optical sensors, enabling fine-grained target perception. Such systems are essential for advancing the practical application of detection algorithms and improving maritime target surveillance capabilities. To this end, this study constructs a maritime multimodal data resource system using multisource data collected from the sea area near a port in the Bohai Sea. Data were acquired using SAR, radar, visible-light cameras, infrared cameras, and other sensors mounted on shore-based and airborne platforms. The data were labeled by performing automatic correlation registration and manual correction. According to the requirements of different tasks, multiple task-oriented multimodal associated datasets were compiled. This paper focuses on one subset of the overall resource system, namely the Dual-Modal Ship Detection, which consists exclusively of visible-light and infrared image pairs. The dataset contains 2163 registered image pairs, with intermodal alignment achieved through an affine transformation. All images were collected in real maritime environments and cover diverse sea conditions and backgrounds, including cloud, rain, fog, and backlighting. The dataset was evaluated using representative algorithms, including YOLO and CFT. Experimental results show that the dataset achieves an mAP@50 of approximately 0.65 with YOLOv8 and 0.63 with CFT, demonstrating its effectiveness in supporting research on optimizing bimodal fusion strategies and enhancing detection robustness in complex maritime scenarios.
A maritime multimodal data resource system provides a foundation for multisensor collaborative detection using radar, Synthetic Aperture Radar (SAR), and electro-optical sensors, enabling fine-grained target perception. Such systems are essential for advancing the practical application of detection algorithms and improving maritime target surveillance capabilities. To this end, this study constructs a maritime multimodal data resource system using multisource data collected from the sea area near a port in the Bohai Sea. Data were acquired using SAR, radar, visible-light cameras, infrared cameras, and other sensors mounted on shore-based and airborne platforms. The data were labeled by performing automatic correlation registration and manual correction. According to the requirements of different tasks, multiple task-oriented multimodal associated datasets were compiled. This paper focuses on one subset of the overall resource system, namely the Dual-Modal Ship Detection, which consists exclusively of visible-light and infrared image pairs. The dataset contains 2163 registered image pairs, with intermodal alignment achieved through an affine transformation. All images were collected in real maritime environments and cover diverse sea conditions and backgrounds, including cloud, rain, fog, and backlighting. The dataset was evaluated using representative algorithms, including YOLO and CFT. Experimental results show that the dataset achieves an mAP@50 of approximately 0.65 with YOLOv8 and 0.63 with CFT, demonstrating its effectiveness in supporting research on optimizing bimodal fusion strategies and enhancing detection robustness in complex maritime scenarios.
The millimeter-Wave (mmWave) radar is widely used in security screening, nondestructive testing, and through-the-wall imaging due to its compact size, high resolution, and strong penetration capability. High-resolution mmWave radar imaging typically requires synthetic aperture emulation, which involves dense two-dimensional spatial sampling via structured scanning on a mechanical platform. However, this process is time-consuming in practical applications. Therefore, many existing studies have focused on reconstructing echo data under sparse sampling conditions for imaging. However, most existing sparse recovery methods assume uniformly random sparse sampling or involve high computational complexity, making them difficult to apply in practical Synthetic Aperture Radar (SAR) imaging systems. This paper proposes a fast, structured sparse, mmWave three-Dimensional (3D) SAR imaging algorithm based on low-rank and smooth Matrix Completion (MC) to address this problem. First, the global low-rank property and local smoothness prior of echo data are analyzed within the framework of near-field mmWave SAR imaging theory. Our analysis demonstrated that structured sparse SAR data arising from missing entire rows or columns in practical scanning can be recovered. Building on this, an MC model incorporating low-rank and smoothness constraints was constructed. This MC model jointly regularizes with nuclear norm and total variation and can be solved efficiently using the Alternating Direction Method of Multipliers (ADMM). Finally, the performance of the proposed algorithm was validated through multiple simulation runs and real-world experiments. Experimental results showed that, using only 20%–30% of randomly sampled rows or columns of echo data, the proposed algorithm can achieve fast data recovery and high-resolution 3D imaging within tens of seconds.
The millimeter-Wave (mmWave) radar is widely used in security screening, nondestructive testing, and through-the-wall imaging due to its compact size, high resolution, and strong penetration capability. High-resolution mmWave radar imaging typically requires synthetic aperture emulation, which involves dense two-dimensional spatial sampling via structured scanning on a mechanical platform. However, this process is time-consuming in practical applications. Therefore, many existing studies have focused on reconstructing echo data under sparse sampling conditions for imaging. However, most existing sparse recovery methods assume uniformly random sparse sampling or involve high computational complexity, making them difficult to apply in practical Synthetic Aperture Radar (SAR) imaging systems. This paper proposes a fast, structured sparse, mmWave three-Dimensional (3D) SAR imaging algorithm based on low-rank and smooth Matrix Completion (MC) to address this problem. First, the global low-rank property and local smoothness prior of echo data are analyzed within the framework of near-field mmWave SAR imaging theory. Our analysis demonstrated that structured sparse SAR data arising from missing entire rows or columns in practical scanning can be recovered. Building on this, an MC model incorporating low-rank and smoothness constraints was constructed. This MC model jointly regularizes with nuclear norm and total variation and can be solved efficiently using the Alternating Direction Method of Multipliers (ADMM). Finally, the performance of the proposed algorithm was validated through multiple simulation runs and real-world experiments. Experimental results showed that, using only 20%–30% of randomly sampled rows or columns of echo data, the proposed algorithm can achieve fast data recovery and high-resolution 3D imaging within tens of seconds.
Small unmanned aerial vehicle (UAV)–borne distributed tomographic synthetic aperture radar (TomoSAR) systems exhibit remarkable residual time-varying baseline errors due to the limited precision of the position and orientation system on small UAV platforms. These errors critically degrade the performance of three-dimensional (3D) target reconstruction. Compared with airborne repeat-pass 3D synthetic aperture radar (SAR), distributed TomoSAR mounted on small UAVs imposes stricter compensation accuracy requirements for time-varying baseline errors because of the altitude constraints of the carrying platform. Under the conditions of low signal-to-noise ratio and substantial time-varying baseline errors, existing estimation methods often fail to provide stable and reliable results. In this paper, a two-step time-varying baseline error estimation method based on image azimuth displacement is proposed. The method sequentially estimates the low-frequency component through the co-registration of the master and slave images and the high-frequency component using a multisquint algorithm. Iterative refinement is applied to enhance estimation accuracy. The experimental results obtained from real C-band small UAV-borne distributed TomoSAR data demonstrate that, compared with the enhanced multisquint processing method, the proposed method considerably reduces the root mean square of differential interferometric phases across most channels, thereby effectively improving interchannel coherence. In addition, the elevation-direction standard deviation of the reconstructed point cloud is reduced from 5.16 to 1.33 m, and the height reconstruction error of building targets is less than 0.5 m, validating the effectiveness and superiority of the proposed method.
Small unmanned aerial vehicle (UAV)–borne distributed tomographic synthetic aperture radar (TomoSAR) systems exhibit remarkable residual time-varying baseline errors due to the limited precision of the position and orientation system on small UAV platforms. These errors critically degrade the performance of three-dimensional (3D) target reconstruction. Compared with airborne repeat-pass 3D synthetic aperture radar (SAR), distributed TomoSAR mounted on small UAVs imposes stricter compensation accuracy requirements for time-varying baseline errors because of the altitude constraints of the carrying platform. Under the conditions of low signal-to-noise ratio and substantial time-varying baseline errors, existing estimation methods often fail to provide stable and reliable results. In this paper, a two-step time-varying baseline error estimation method based on image azimuth displacement is proposed. The method sequentially estimates the low-frequency component through the co-registration of the master and slave images and the high-frequency component using a multisquint algorithm. Iterative refinement is applied to enhance estimation accuracy. The experimental results obtained from real C-band small UAV-borne distributed TomoSAR data demonstrate that, compared with the enhanced multisquint processing method, the proposed method considerably reduces the root mean square of differential interferometric phases across most channels, thereby effectively improving interchannel coherence. In addition, the elevation-direction standard deviation of the reconstructed point cloud is reduced from 5.16 to 1.33 m, and the height reconstruction error of building targets is less than 0.5 m, validating the effectiveness and superiority of the proposed method.
Pulse Doppler radar provides all-weather operational capability and enables simultaneous acquisition of target range and velocity through Range-Doppler (RD) maps. In near-vertical flight scenarios, the geometric structure of RD maps implicitly encodes key platform motion parameters, including altitude, velocity, and pitch angle. However, these parameters are strongly coupled in the RD domain, making effective decoupling difficult for traditional signal-processing-based inversion methods, particularly under complex terrain and near-vertical incidence conditions. Although recent advances in deep learning have shown strong potential for motion information sensing, multitask learning in this context still faces challenges in achieving both real-time performance and high estimation accuracy. To address these issues, this study proposes a novel network architecture, termed Range-Doppler Map Fusion Network (RDMFNet), that performs multirepresentation information fusion via shared encoders and parallel decoders, along with a two-stage progressive training strategy to enhance parameter estimation accuracy. Experimental results show that RDMFNet achieves estimation errors of 14.447 m for altitude, 4.635 m/s for velocity, and 0.755° for pitch angle, demonstrating its effectiveness for high-precision, real-time perception.
Pulse Doppler radar provides all-weather operational capability and enables simultaneous acquisition of target range and velocity through Range-Doppler (RD) maps. In near-vertical flight scenarios, the geometric structure of RD maps implicitly encodes key platform motion parameters, including altitude, velocity, and pitch angle. However, these parameters are strongly coupled in the RD domain, making effective decoupling difficult for traditional signal-processing-based inversion methods, particularly under complex terrain and near-vertical incidence conditions. Although recent advances in deep learning have shown strong potential for motion information sensing, multitask learning in this context still faces challenges in achieving both real-time performance and high estimation accuracy. To address these issues, this study proposes a novel network architecture, termed Range-Doppler Map Fusion Network (RDMFNet), that performs multirepresentation information fusion via shared encoders and parallel decoders, along with a two-stage progressive training strategy to enhance parameter estimation accuracy. Experimental results show that RDMFNet achieves estimation errors of 14.447 m for altitude, 4.635 m/s for velocity, and 0.755° for pitch angle, demonstrating its effectiveness for high-precision, real-time perception.
Radar systems acquire target information by transmitting waveforms, receiving echoes, and processing signals; thus, waveform performance is a critical determinant of radar system performance. Compared with other radar systems, Synthetic Aperture Radar (SAR) operates under unique conditions, including distributed target scenes, waveforms with large time-bandwidth products, wide-swath and long-range imaging, and range–azimuth coupling. These characteristics impose additional stringent requirements on SAR waveform design. Drawing on the authors’ research and expertise in SAR waveform coding, this paper reviews recent domestic and international advances in SAR waveform design, discusses key technical challenges, and highlights the role of waveform design in enhancing system imaging performance. Finally, this study outlines future trends and potential directions for SAR waveform design methodologies.
Radar systems acquire target information by transmitting waveforms, receiving echoes, and processing signals; thus, waveform performance is a critical determinant of radar system performance. Compared with other radar systems, Synthetic Aperture Radar (SAR) operates under unique conditions, including distributed target scenes, waveforms with large time-bandwidth products, wide-swath and long-range imaging, and range–azimuth coupling. These characteristics impose additional stringent requirements on SAR waveform design. Drawing on the authors’ research and expertise in SAR waveform coding, this paper reviews recent domestic and international advances in SAR waveform design, discusses key technical challenges, and highlights the role of waveform design in enhancing system imaging performance. Finally, this study outlines future trends and potential directions for SAR waveform design methodologies.
,
Crevasse detection using Ground-Penetrating Radar (GPR) is crucial for glacier and climate studies and for ensuring safety in glacial regions. To address challenges including large texture variability in crevasses in polar environments, high false-alarm rates, limited real-time performance, and poor generalization, this study proposes a domain-adversarial learning–based automatic crevasse detection method that balances high accuracy and real-time efficiency. Using GPR data acquired from diverse regions and complex scenarios, an adversarial learning mechanism is established between a feature extractor and a domain discriminator, enabling the method to maintain discriminative feature extraction while effectively reducing interdomain distribution discrepancies. This leads to cross-domain feature alignment, enhancing robustness and generalization across heterogeneous data sources. In the feature extraction stage, a wavelet-residual-network-based crevasse feature extractor is designed. By introducing a learnable multiscale wavelet convolution module into the first layer of the residual network, the model adaptively extracts multiscale crevasse features from GPR data, significantly enhancing the separability between crevasse regions and continuous snow layers in the feature space. Experiments were conducted on two GPR datasets: the 2015 McMurdo Shear Zone dataset from Antarctica and a Greenland dataset from the Arctic region. The experimental results demonstrate that the proposed model achieves an average detection accuracy of 95.70%, an F1-score of 95.50%, and a false-alarm rate of 1.87%, with an average inference time of 5.26 ms per sample, thereby meeting the requirements for real-time crevasse warning during field GPR acquisition. Overall, the proposed method achieves a favorable balance among high accuracy, low false-alarm rate, and real-time performance across multiscene and cross-regional GPR data, demonstrating its suitability for safety assurance during Antarctic expeditions and for glacier crevasse detection in polar research.
Crevasse detection using Ground-Penetrating Radar (GPR) is crucial for glacier and climate studies and for ensuring safety in glacial regions. To address challenges including large texture variability in crevasses in polar environments, high false-alarm rates, limited real-time performance, and poor generalization, this study proposes a domain-adversarial learning–based automatic crevasse detection method that balances high accuracy and real-time efficiency. Using GPR data acquired from diverse regions and complex scenarios, an adversarial learning mechanism is established between a feature extractor and a domain discriminator, enabling the method to maintain discriminative feature extraction while effectively reducing interdomain distribution discrepancies. This leads to cross-domain feature alignment, enhancing robustness and generalization across heterogeneous data sources. In the feature extraction stage, a wavelet-residual-network-based crevasse feature extractor is designed. By introducing a learnable multiscale wavelet convolution module into the first layer of the residual network, the model adaptively extracts multiscale crevasse features from GPR data, significantly enhancing the separability between crevasse regions and continuous snow layers in the feature space. Experiments were conducted on two GPR datasets: the 2015 McMurdo Shear Zone dataset from Antarctica and a Greenland dataset from the Arctic region. The experimental results demonstrate that the proposed model achieves an average detection accuracy of 95.70%, an F1-score of 95.50%, and a false-alarm rate of 1.87%, with an average inference time of 5.26 ms per sample, thereby meeting the requirements for real-time crevasse warning during field GPR acquisition. Overall, the proposed method achieves a favorable balance among high accuracy, low false-alarm rate, and real-time performance across multiscene and cross-regional GPR data, demonstrating its suitability for safety assurance during Antarctic expeditions and for glacier crevasse detection in polar research.
The automatic target recognition performance of radar is critically dependent on the quality of features extracted from target echo signals. As the information carrier that actively shapes echo signals, the transmitted waveform substantially affects the target classification performance. However, conventional waveform design is often decoupled from classifier optimization, thereby ignoring the critical synergy between the two. This disconnect, combined with the lack of a direct link between waveform optimization criteria and task-specific classification metrics, limits the target classification performance. Most existing approaches are confined to monostatic radar models. Further, they fail to establish relationships between the target’s aspect angle, the transmitted waveform, and classification performance, and lack a cooperative waveform design mechanism among nodes. Hence, they are unable to achieve spatial and waveform diversity gains. To overcome these limitations, this paper proposes an end-to-end “waveform aspect matching” optimization framework for target classification in distributed radar systems. This framework parameterizes the waveform as a trainable waveform generation module, cascaded with a downstream classification network. This transforms the isolated waveform design problem into a joint optimization of the waveform and classifier, directly guided by the classification task. Leveraging prior target information, the model is trained to jointly optimize and produce aspect-matched waveforms along with the corresponding classification network. Furthermore, to enhance the classification performance in distributed radar systems, a dual-branch network based on noncausal state-space duality modules is proposed to extract and fuse multiview information. Experimental results demonstrate that the proposed method can synergistically utilize waveform and spatial diversity to improve the target classification performance. It demonstrates robustness against node failures, offering a novel solution for intelligent waveform design in distributed radar systems.
The automatic target recognition performance of radar is critically dependent on the quality of features extracted from target echo signals. As the information carrier that actively shapes echo signals, the transmitted waveform substantially affects the target classification performance. However, conventional waveform design is often decoupled from classifier optimization, thereby ignoring the critical synergy between the two. This disconnect, combined with the lack of a direct link between waveform optimization criteria and task-specific classification metrics, limits the target classification performance. Most existing approaches are confined to monostatic radar models. Further, they fail to establish relationships between the target’s aspect angle, the transmitted waveform, and classification performance, and lack a cooperative waveform design mechanism among nodes. Hence, they are unable to achieve spatial and waveform diversity gains. To overcome these limitations, this paper proposes an end-to-end “waveform aspect matching” optimization framework for target classification in distributed radar systems. This framework parameterizes the waveform as a trainable waveform generation module, cascaded with a downstream classification network. This transforms the isolated waveform design problem into a joint optimization of the waveform and classifier, directly guided by the classification task. Leveraging prior target information, the model is trained to jointly optimize and produce aspect-matched waveforms along with the corresponding classification network. Furthermore, to enhance the classification performance in distributed radar systems, a dual-branch network based on noncausal state-space duality modules is proposed to extract and fuse multiview information. Experimental results demonstrate that the proposed method can synergistically utilize waveform and spatial diversity to improve the target classification performance. It demonstrates robustness against node failures, offering a novel solution for intelligent waveform design in distributed radar systems.
,
A large-scale Vision-Language Model (VLM) pre-trained on massive image–text datasets performs well when processing natural images. However, there are two major challenges in applying it to Synthetic Aperture Radar (SAR) images: (1) the high cost of high-quality text annotation limits the construction of SAR image–text paired datasets, and (2) the considerable differences in visual characteristics between SAR images and optical natural images increase the difficulty of cross-domain knowledge transfer. To address these problems, this study developed a knowledge transfer method for VLM tailored to SAR images. First, this study leveraged paired SAR and optical remote sensing images and employed a generative VLM to automatically produce textual descriptions of the optical images, thereby indirectly constructing a low-cost SAR–text paired dataset. Second, a two-stage transfer strategy was designed to address the large domain discrepancy between natural and SAR images, reducing the difficulty of each transfer stage. Finally, experimental validation was conducted through the zero-shot scene classification, image retrieval, and object recognition of SAR images. The results demonstrated that the proposed method enables effective knowledge transfer from a large-scale VLM to the SAR image domain.
A large-scale Vision-Language Model (VLM) pre-trained on massive image–text datasets performs well when processing natural images. However, there are two major challenges in applying it to Synthetic Aperture Radar (SAR) images: (1) the high cost of high-quality text annotation limits the construction of SAR image–text paired datasets, and (2) the considerable differences in visual characteristics between SAR images and optical natural images increase the difficulty of cross-domain knowledge transfer. To address these problems, this study developed a knowledge transfer method for VLM tailored to SAR images. First, this study leveraged paired SAR and optical remote sensing images and employed a generative VLM to automatically produce textual descriptions of the optical images, thereby indirectly constructing a low-cost SAR–text paired dataset. Second, a two-stage transfer strategy was designed to address the large domain discrepancy between natural and SAR images, reducing the difficulty of each transfer stage. Finally, experimental validation was conducted through the zero-shot scene classification, image retrieval, and object recognition of SAR images. The results demonstrated that the proposed method enables effective knowledge transfer from a large-scale VLM to the SAR image domain.
Reinforcement Learning (RL) is a critical approach for enabling cognitive radar target detection. Existing studies primarily focus on detection methods for centralized Multiple-Input Multiple-Output (MIMO) radar, which are limited to a single observation perspective. To address this issue, this paper proposes an RL-based multi-target detection method for a distributed MIMO radar system that possesses waveform and spatial diversity. The proposed method exploits spatial diversity to ensure robust target detection, while waveform diversity is used to construct a Markov decision process. Specifically, the radar first perceives target attributes through statistical signal detection techniques, then optimizes the transmit waveform accordingly, and iteratively updates its understanding of the environmental context using accumulated experience. This cyclic process gradually converges, yielding radar waveforms focused on target directions and achieving improved detection performance. To facilitate target localization, a maximization grid-based generalized likelihood ratio test detector for multi-antenna configurations is derived, using regularly shaped grids as the cell under test. For waveform optimization, two types of optimization problems, namely conventional and strong-target-limited formulations, are developed, and their solutions are obtained using continuous convex approximation. Simulation results across static and dynamic scenarios demonstrate that the proposed method can autonomously perceive environmental context and achieve superior detection performance compared with benchmark methods, particularly in weak target detection.
Reinforcement Learning (RL) is a critical approach for enabling cognitive radar target detection. Existing studies primarily focus on detection methods for centralized Multiple-Input Multiple-Output (MIMO) radar, which are limited to a single observation perspective. To address this issue, this paper proposes an RL-based multi-target detection method for a distributed MIMO radar system that possesses waveform and spatial diversity. The proposed method exploits spatial diversity to ensure robust target detection, while waveform diversity is used to construct a Markov decision process. Specifically, the radar first perceives target attributes through statistical signal detection techniques, then optimizes the transmit waveform accordingly, and iteratively updates its understanding of the environmental context using accumulated experience. This cyclic process gradually converges, yielding radar waveforms focused on target directions and achieving improved detection performance. To facilitate target localization, a maximization grid-based generalized likelihood ratio test detector for multi-antenna configurations is derived, using regularly shaped grids as the cell under test. For waveform optimization, two types of optimization problems, namely conventional and strong-target-limited formulations, are developed, and their solutions are obtained using continuous convex approximation. Simulation results across static and dynamic scenarios demonstrate that the proposed method can autonomously perceive environmental context and achieve superior detection performance compared with benchmark methods, particularly in weak target detection.
Vortex electromagnetic waves carrying Orbital Angular Momentum (OAM) can meet the requirements of modern radar detection systems for high resolution and precision. However, existing OAM beam generation methods suffer from limitations, such as insufficient multimodal purity and strong mutual coupling between array elements. To overcome these challenges, this paper designs and optimizes pyramidal horn antenna elements based on a uniform concentric circular array, thereby establishing a multimodal OAM array model. A double-layer metal ground plane structure is introduced to effectively suppress mutual coupling between array elements. Furthermore, the array configuration is optimized to generate high-purity, co-directional multimodal OAM beams. On this basis, a genetic algorithm is used to further optimize the design for generating multimodal OAM beams with low sidelobes. Full-wave simulations show that the optimized array achieves an active reflection coefficient below −10 dB, indicating substantially suppression of the mutual coupling between elements. The proposed array exhibits a stable structure suitable for engineering applications and supports the generation of 14 co-directional OAM beams with modal purity exceeding 0.92 and sidelobes below −13 dB. Finally, the performance of the designed array is validated through fabrication, testing, and super-resolution imaging experiments.
Vortex electromagnetic waves carrying Orbital Angular Momentum (OAM) can meet the requirements of modern radar detection systems for high resolution and precision. However, existing OAM beam generation methods suffer from limitations, such as insufficient multimodal purity and strong mutual coupling between array elements. To overcome these challenges, this paper designs and optimizes pyramidal horn antenna elements based on a uniform concentric circular array, thereby establishing a multimodal OAM array model. A double-layer metal ground plane structure is introduced to effectively suppress mutual coupling between array elements. Furthermore, the array configuration is optimized to generate high-purity, co-directional multimodal OAM beams. On this basis, a genetic algorithm is used to further optimize the design for generating multimodal OAM beams with low sidelobes. Full-wave simulations show that the optimized array achieves an active reflection coefficient below −10 dB, indicating substantially suppression of the mutual coupling between elements. The proposed array exhibits a stable structure suitable for engineering applications and supports the generation of 14 co-directional OAM beams with modal purity exceeding 0.92 and sidelobes below −13 dB. Finally, the performance of the designed array is validated through fabrication, testing, and super-resolution imaging experiments.
Coherent Frequency Diverse Array (FDA) radar demonstrates significant potential for wide-area search tasks due to its simple system architecture, flexible beam scanning, and high transmit Degrees of Freedom (DOF). However, its inherent beam-scanning mechanism reduces dwell time in specific directions, thereby limiting the imaging range resolution when a conventional wideband waveform is used. To resolve the intrinsic contradiction between wide-area search and high-resolution imaging, this paper proposes a deep learning-based integrated search-imaging waveform design method. By leveraging the multi-DoF flexible transmission capability of coherent FDA, the proposed method customizes multidimensional transmit resources, including waveform, bandwidth, and transmit gain, for multiple Regions of Interest (ROIs) while preserving wide-coverage search performance. To address the nonconvex optimization problem with dual constraints of constant modulus and low correlation in baseband waveform design, a residual autoencoder-based optimizer is developed. This network directly learns and establishes a high-dimensional nonlinear mapping from the initial phase space to the optimized phase space that satisfies predefined performance criteria. The network efficiently generates a set of phase-coded subwaveforms exhibiting low autocorrelation sidelobes and low cross-correlation levels for multiple ROIs. Simulation results validate the effectiveness of this method, demonstrating that the designed waveforms achieve higher processing gain (compared with the narrowband searching mode) and improved imaging resolution in the designated ROIs during simultaneous search and multitarget imaging. Moreover, the autocorrelation and cross-correlation performance of the proposed method significantly outperforms that of conventional approaches, indicating that it provides an effective solution for enhancing the multitask capabilities of modern radar systems.
Coherent Frequency Diverse Array (FDA) radar demonstrates significant potential for wide-area search tasks due to its simple system architecture, flexible beam scanning, and high transmit Degrees of Freedom (DOF). However, its inherent beam-scanning mechanism reduces dwell time in specific directions, thereby limiting the imaging range resolution when a conventional wideband waveform is used. To resolve the intrinsic contradiction between wide-area search and high-resolution imaging, this paper proposes a deep learning-based integrated search-imaging waveform design method. By leveraging the multi-DoF flexible transmission capability of coherent FDA, the proposed method customizes multidimensional transmit resources, including waveform, bandwidth, and transmit gain, for multiple Regions of Interest (ROIs) while preserving wide-coverage search performance. To address the nonconvex optimization problem with dual constraints of constant modulus and low correlation in baseband waveform design, a residual autoencoder-based optimizer is developed. This network directly learns and establishes a high-dimensional nonlinear mapping from the initial phase space to the optimized phase space that satisfies predefined performance criteria. The network efficiently generates a set of phase-coded subwaveforms exhibiting low autocorrelation sidelobes and low cross-correlation levels for multiple ROIs. Simulation results validate the effectiveness of this method, demonstrating that the designed waveforms achieve higher processing gain (compared with the narrowband searching mode) and improved imaging resolution in the designated ROIs during simultaneous search and multitarget imaging. Moreover, the autocorrelation and cross-correlation performance of the proposed method significantly outperforms that of conventional approaches, indicating that it provides an effective solution for enhancing the multitask capabilities of modern radar systems.
By applying phase coding in transmit elements and pulses, the Element-Pulse Coding Multiple-Input Multiple-Output (EPC-MIMO) radar can effectively suppress mainlobe deceptive interference. However, this approach remains ineffective against mainlobe blanket interference. To address this drawback, this paper investigates the mainlobe blanket interference suppression using a Polarization Element-Pulse Coding Multiple-Input Multiple-Output (PEPC-MIMO) radar system. Specifically, within the framework of stable principal component pursuit decomposition, the interference suppression problem is formulated as a “low-rank + sparse” optimization model by exploiting the low-rank structure of the received signal in the joint time-space-polarization domain. The resulting optimization problem is solved iteratively using a Limited-memory Broyden-Fletcher-Goldfarb-Shanno-based Alternating Optimization (L-BFGS-AO) algorithm, thereby enabling accurate separation of target echoes from mainlobe blanket interference. Furthermore, a sparse reconstruction-based parameter estimation method is proposed to estimate the target’s transmit angle, receive angle, and range ambiguity region. These estimates are then used to construct optimal receive weight vectors for the weighted summation of signals across channels. Simulation results demonstrate the effectiveness of the proposed approach in suppressing mainlobe blanket interference without requiring prior knowledge of the interference.
By applying phase coding in transmit elements and pulses, the Element-Pulse Coding Multiple-Input Multiple-Output (EPC-MIMO) radar can effectively suppress mainlobe deceptive interference. However, this approach remains ineffective against mainlobe blanket interference. To address this drawback, this paper investigates the mainlobe blanket interference suppression using a Polarization Element-Pulse Coding Multiple-Input Multiple-Output (PEPC-MIMO) radar system. Specifically, within the framework of stable principal component pursuit decomposition, the interference suppression problem is formulated as a “low-rank + sparse” optimization model by exploiting the low-rank structure of the received signal in the joint time-space-polarization domain. The resulting optimization problem is solved iteratively using a Limited-memory Broyden-Fletcher-Goldfarb-Shanno-based Alternating Optimization (L-BFGS-AO) algorithm, thereby enabling accurate separation of target echoes from mainlobe blanket interference. Furthermore, a sparse reconstruction-based parameter estimation method is proposed to estimate the target’s transmit angle, receive angle, and range ambiguity region. These estimates are then used to construct optimal receive weight vectors for the weighted summation of signals across channels. Simulation results demonstrate the effectiveness of the proposed approach in suppressing mainlobe blanket interference without requiring prior knowledge of the interference.
This study addresses time-frequency synchronization errors in distributed Multiple-Input Multiple-Output (MIMO) radar systems and proposes a joint estimation method for target parameters and system time-frequency biases based on multitemporal measurement data. The method overcomes the limitations of traditional approaches that rely on singletemporal measurement data and direct-path signals, enabling high-accuracy joint parameter estimation through multiepoch data fusion without requiring direct-path information. The proposed method adopts a two-step strategy that combines a closed-form solution with iterative optimization. First, a closed-form solution is derived within a two-stage weighted least-squares framework using only the first- and last-epoch observations to obtain initial estimates of the target position, velocity, and auxiliary variables. This stage explicitly models second-order error terms and optimizes the construction of the weighting matrix, significantly improving accuracy and robustness under high-error conditions. Second, using the closed-form estimates as initialization, a maximum likelihood-maximum a posteriori objective function is formulated based on the full multi-epoch measurement data, and a trust-region iterative optimization method is applied to refine the estimates and recover the time-frequency bias parameters. Simulation results show that the proposed method outperforms existing approaches across various error levels and geometric configurations, significantly enhancing the accuracy and robustness of target localization, velocity estimation, and time-frequency bias estimation. These results demonstrate strong theoretical significance and promising practical application potential.
This study addresses time-frequency synchronization errors in distributed Multiple-Input Multiple-Output (MIMO) radar systems and proposes a joint estimation method for target parameters and system time-frequency biases based on multitemporal measurement data. The method overcomes the limitations of traditional approaches that rely on singletemporal measurement data and direct-path signals, enabling high-accuracy joint parameter estimation through multiepoch data fusion without requiring direct-path information. The proposed method adopts a two-step strategy that combines a closed-form solution with iterative optimization. First, a closed-form solution is derived within a two-stage weighted least-squares framework using only the first- and last-epoch observations to obtain initial estimates of the target position, velocity, and auxiliary variables. This stage explicitly models second-order error terms and optimizes the construction of the weighting matrix, significantly improving accuracy and robustness under high-error conditions. Second, using the closed-form estimates as initialization, a maximum likelihood-maximum a posteriori objective function is formulated based on the full multi-epoch measurement data, and a trust-region iterative optimization method is applied to refine the estimates and recover the time-frequency bias parameters. Simulation results show that the proposed method outperforms existing approaches across various error levels and geometric configurations, significantly enhancing the accuracy and robustness of target localization, velocity estimation, and time-frequency bias estimation. These results demonstrate strong theoretical significance and promising practical application potential.
Space-Time Adaptive Processing (STAP) is a key technique used for ground/sea clutter suppression and moving target detection in airborne radar. However, under range ambiguity conditions, the inherent clutter nonstationarity in airborne bistatic radar violates the independent and identically distributed assumption required for training samples, significantly degrading the performance of conventional STAP methods. To address this issue, this study first analyzed the limitations of the Conventional Beamforming (CBF)-based disambiguation approach, which exhibits a trade-off between mainlobe gain loss and sidelobe suppression. A method based on a cascaded blocking matrix and adaptive beamforming is proposed for separating range-ambiguous clutter. Although this method improves upon the CBF-based approach, it introduces noise distortion, which limits further improvements in clutter separation and suppression accuracy. Hence, a novel method based on beam pattern reconstruction is proposed to overcome this drawback. This method formulates an optimization problem incorporating beam-maintenance and sidelobe-control terms to design spatial filter weights, effectively separating range-ambiguous clutter while preserving the mainlobe target gain and suppressing sidelobe clutter. Subsequently, angle-Doppler compensation is applied to the separated clutter, followed by final suppression using a subarray-based STAP method incorporating a joint three-channel Doppler transform. Simulation results revealed that compared with typical methods, the proposed approach more effectively separated clutter, significantly narrowed the mainlobe notch width, and limited the output signal-to-clutter-plus-noise ratio loss to <3 dB, thereby markedly enhancing clutter suppression performance and target detection capability under range ambiguity conditions.
Space-Time Adaptive Processing (STAP) is a key technique used for ground/sea clutter suppression and moving target detection in airborne radar. However, under range ambiguity conditions, the inherent clutter nonstationarity in airborne bistatic radar violates the independent and identically distributed assumption required for training samples, significantly degrading the performance of conventional STAP methods. To address this issue, this study first analyzed the limitations of the Conventional Beamforming (CBF)-based disambiguation approach, which exhibits a trade-off between mainlobe gain loss and sidelobe suppression. A method based on a cascaded blocking matrix and adaptive beamforming is proposed for separating range-ambiguous clutter. Although this method improves upon the CBF-based approach, it introduces noise distortion, which limits further improvements in clutter separation and suppression accuracy. Hence, a novel method based on beam pattern reconstruction is proposed to overcome this drawback. This method formulates an optimization problem incorporating beam-maintenance and sidelobe-control terms to design spatial filter weights, effectively separating range-ambiguous clutter while preserving the mainlobe target gain and suppressing sidelobe clutter. Subsequently, angle-Doppler compensation is applied to the separated clutter, followed by final suppression using a subarray-based STAP method incorporating a joint three-channel Doppler transform. Simulation results revealed that compared with typical methods, the proposed approach more effectively separated clutter, significantly narrowed the mainlobe notch width, and limited the output signal-to-clutter-plus-noise ratio loss to <3 dB, thereby markedly enhancing clutter suppression performance and target detection capability under range ambiguity conditions.
Ocean currents play a critical role in global climate regulation. Synthetic Aperture Radar (SAR) provides high-resolution observational support for ocean current detection by measuring Doppler shifts; however, SAR Doppler shifts contain multiple contributing components. To accurately retrieve ocean currents from these data, nongeophysical contributions must be precisely corrected, and wind- and wave-induced Doppler shifts must be accurately estimated. This paper proposes a machine learning-based method for modeling such shifts and retrieving ocean currents from Sentinel-1 SAR data. First, nongeophysical contributions in the SAR Doppler shift are precisely corrected to remove the effects unrelated to ocean motion. Second, BackproPagation Neural Network (BPNN) and eXtreme Gradient Boosting (XGBoost) models, optimized using the particle swarm optimization algorithm, are developed to describe the nonlinear relationship between the wind-wave Doppler shift and sea-surface wind-wave parameters derived from SAR data. Finally, the corrected Doppler shift is utilized to retrieve ocean surface current velocities. This paper comparatively evaluates the estimation accuracies of the wind-wave Doppler shifts obtained using the BPNN and XGBoost models, as well as the respective influence of each model's performance on the effectiveness of ocean current retrieval. Results indicate that the XGBoost model achieves superior estimation accuracy compared with the BPNN model. The Root Mean Square Error (RMSE) of the Doppler shift estimated by the XGBoost model is approximately 4.043 Hz, which is 2.898 Hz lower than that of the BPNN model. Compared with those of the HYCOM current data, the RMSE of the currents retrieved by the XGBoost model is about 0.202 m/s; this value is reduced by 0.122 m/s compared with that of the BPNN model. Validations against the current velocities detected by HF radar show that the RMSE of currents retrieved by the XGBoost model is 0.21 m/s, representing a 16% reduction compared with that of the BPNN model. These findings indicate that the proposed technical approach for ocean current retrieval using spaceborne SAR is highly accurate.
Ocean currents play a critical role in global climate regulation. Synthetic Aperture Radar (SAR) provides high-resolution observational support for ocean current detection by measuring Doppler shifts; however, SAR Doppler shifts contain multiple contributing components. To accurately retrieve ocean currents from these data, nongeophysical contributions must be precisely corrected, and wind- and wave-induced Doppler shifts must be accurately estimated. This paper proposes a machine learning-based method for modeling such shifts and retrieving ocean currents from Sentinel-1 SAR data. First, nongeophysical contributions in the SAR Doppler shift are precisely corrected to remove the effects unrelated to ocean motion. Second, BackproPagation Neural Network (BPNN) and eXtreme Gradient Boosting (XGBoost) models, optimized using the particle swarm optimization algorithm, are developed to describe the nonlinear relationship between the wind-wave Doppler shift and sea-surface wind-wave parameters derived from SAR data. Finally, the corrected Doppler shift is utilized to retrieve ocean surface current velocities. This paper comparatively evaluates the estimation accuracies of the wind-wave Doppler shifts obtained using the BPNN and XGBoost models, as well as the respective influence of each model's performance on the effectiveness of ocean current retrieval. Results indicate that the XGBoost model achieves superior estimation accuracy compared with the BPNN model. The Root Mean Square Error (RMSE) of the Doppler shift estimated by the XGBoost model is approximately 4.043 Hz, which is 2.898 Hz lower than that of the BPNN model. Compared with those of the HYCOM current data, the RMSE of the currents retrieved by the XGBoost model is about 0.202 m/s; this value is reduced by 0.122 m/s compared with that of the BPNN model. Validations against the current velocities detected by HF radar show that the RMSE of currents retrieved by the XGBoost model is 0.21 m/s, representing a 16% reduction compared with that of the BPNN model. These findings indicate that the proposed technical approach for ocean current retrieval using spaceborne SAR is highly accurate.
Direction of Arrival (DOA) estimation for low-elevation angle targets is a critical challenge in meter-wave and holographic staring radar systems, as its accuracy directly affects target height measurement performance. Traditional beamspace methods reduce computational complexity by projecting high-dimensional element-space data onto a low-dimensional beamspace using a beamformer. However, this lossy mapping leads to partial information loss, resulting in degraded elevation-angle estimation accuracy compared to that of element-space methods. To address this issue, this study proposes a high-accuracy beamspace DOA estimation method for low-elevation angle targets. First, the Cramér-Rao Bound (CRB) for both element-space and beamspace DOA estimation is derived, and the conditions under which these bounds are equal are analyzed. Since these conditions are difficult to satisfy in practical scenarios, an approximate-condition-based beamformer design strategy is developed to reduce data dimensionality while preserving effective target information. Finally, precise elevation-angle estimation is achieved using the maximum likelihood criterion. Simulation and experimental results show that the proposed method significantly reduces data dimensionality while maintaining estimation accuracy comparable to that of element-space methods at low-elevation angles, clearly outperforming existing beamspace algorithms.
Direction of Arrival (DOA) estimation for low-elevation angle targets is a critical challenge in meter-wave and holographic staring radar systems, as its accuracy directly affects target height measurement performance. Traditional beamspace methods reduce computational complexity by projecting high-dimensional element-space data onto a low-dimensional beamspace using a beamformer. However, this lossy mapping leads to partial information loss, resulting in degraded elevation-angle estimation accuracy compared to that of element-space methods. To address this issue, this study proposes a high-accuracy beamspace DOA estimation method for low-elevation angle targets. First, the Cramér-Rao Bound (CRB) for both element-space and beamspace DOA estimation is derived, and the conditions under which these bounds are equal are analyzed. Since these conditions are difficult to satisfy in practical scenarios, an approximate-condition-based beamformer design strategy is developed to reduce data dimensionality while preserving effective target information. Finally, precise elevation-angle estimation is achieved using the maximum likelihood criterion. Simulation and experimental results show that the proposed method significantly reduces data dimensionality while maintaining estimation accuracy comparable to that of element-space methods at low-elevation angles, clearly outperforming existing beamspace algorithms.
Deep learning is primarily used for target detection in Synthetic Aperture Radar (SAR) images; however, its performance heavily relies on large-scale labeled datasets. The detection performance of deep learning models degrades when applied to SAR data with varying distributions, hindering their real-world applicability. In addition, manual labeling of SAR data is costly. Hence, cross-domain learning strategies based on multisource information are being explored to address these challenges. These strategies can assist detection models in realizing cross-domain knowledge migration by integrating prior information from optical remote sensing images or heterogeneous SAR images acquired from different sensors. This paper focuses on cross-domain learning technologies within the deep learning framework. In addition, it provides a systematic overview of the latest research progress in this field and analyzes the core issues, advantages, and applicable scenarios of existing technologies from a methodological perspective. It outlines future research directions based on the law of technological evolution, aiming to offer theoretical support and methodological references to enhance the generalizability of target detection in SAR images.
Deep learning is primarily used for target detection in Synthetic Aperture Radar (SAR) images; however, its performance heavily relies on large-scale labeled datasets. The detection performance of deep learning models degrades when applied to SAR data with varying distributions, hindering their real-world applicability. In addition, manual labeling of SAR data is costly. Hence, cross-domain learning strategies based on multisource information are being explored to address these challenges. These strategies can assist detection models in realizing cross-domain knowledge migration by integrating prior information from optical remote sensing images or heterogeneous SAR images acquired from different sensors. This paper focuses on cross-domain learning technologies within the deep learning framework. In addition, it provides a systematic overview of the latest research progress in this field and analyzes the core issues, advantages, and applicable scenarios of existing technologies from a methodological perspective. It outlines future research directions based on the law of technological evolution, aiming to offer theoretical support and methodological references to enhance the generalizability of target detection in SAR images.
Synthetic Aperture Radar (SAR) plays a pivotal role in military reconnaissance and remote-sensing applications, given its all-weather, day-and-night operability and high-resolution imaging performance. However, diverse jamming techniques in modern complex electromagnetic environments severely distort SAR echo signals, leading to blurred or distorted imaging results and, in extreme cases, complete target unrecognizability. Given the fundamental differences in formation mechanisms and suppression strategies of different jamming types, precise jamming identification is a core prerequisite for effective counterjamming. Current SAR jamming identification methods face two major challenges. First, when the energy of the jamming signal is comparable to that of the target signal, the jamming features are easily masked, making reliable detection and identification difficult. Second, existing identification networks generally suffer from excessive complexity and poor real-time performance, limiting their practicality in engineering applications. To address these issues, this paper proposes a lightweight network-based non-spoofing active jamming identification method for SAR under low Jamming-to-Signal Ratio (JSR) conditions. This method introduces two key components: a lattice transform block that boosts interference discrimination at low JSR by refining fine-grained feature extraction and a hyperkernel-aware module that, through a custom hyperkernel block based on point target imaging, enhances context capture while ensuring algorithmic lightweighting. The superiority of the proposed method is validated through multidimensional evaluations, including effectiveness analyses of the modules, accuracy-complexity trade-off analysis of different models, and robustness testing under varying JSR conditions. The proposed method maintains high identification performance even under low JSR conditions while meeting real-time computational efficiency requirements.
Synthetic Aperture Radar (SAR) plays a pivotal role in military reconnaissance and remote-sensing applications, given its all-weather, day-and-night operability and high-resolution imaging performance. However, diverse jamming techniques in modern complex electromagnetic environments severely distort SAR echo signals, leading to blurred or distorted imaging results and, in extreme cases, complete target unrecognizability. Given the fundamental differences in formation mechanisms and suppression strategies of different jamming types, precise jamming identification is a core prerequisite for effective counterjamming. Current SAR jamming identification methods face two major challenges. First, when the energy of the jamming signal is comparable to that of the target signal, the jamming features are easily masked, making reliable detection and identification difficult. Second, existing identification networks generally suffer from excessive complexity and poor real-time performance, limiting their practicality in engineering applications. To address these issues, this paper proposes a lightweight network-based non-spoofing active jamming identification method for SAR under low Jamming-to-Signal Ratio (JSR) conditions. This method introduces two key components: a lattice transform block that boosts interference discrimination at low JSR by refining fine-grained feature extraction and a hyperkernel-aware module that, through a custom hyperkernel block based on point target imaging, enhances context capture while ensuring algorithmic lightweighting. The superiority of the proposed method is validated through multidimensional evaluations, including effectiveness analyses of the modules, accuracy-complexity trade-off analysis of different models, and robustness testing under varying JSR conditions. The proposed method maintains high identification performance even under low JSR conditions while meeting real-time computational efficiency requirements.
Chang’e-7 will carry a fully polarimetric Synthetic Aperture Radar (SAR) to investigate the topography and material properties of the lunar polar regions, which necessitates reliable polarimetric calibration. However, conventional ground-based calibration strategies are infeasible for lunar missions, underscoring the need for new relative polarimetric calibration methods tailored to the lunar surface. To address this challenge, we adopt a normal-incidence observation geometry and analyze how the co-polarization ratio \begin{document}$ {\sigma _{{\text{HH}}}} $\end{document} \begin{document}$ {\sigma _{{\text{VV}}}} $\end{document} \begin{document}$ \sigma $\end{document} \begin{document}$ {\sigma _{{\text{HH}}}} $\end{document} \begin{document}$ {\sigma _{{\text{VV}}}} $\end{document} \begin{document}$ \sigma $\end{document}
Synthetic Aperture Radar (SAR) and optical imagery are two key remote-sensing modalities in Earth observation, and cross-modal image matching between them is widely applied in tasks such as image fusion, joint interpretation, and high-precision geolocation. In recent years, with the rapid growth of Earth-observation data, the importance of cross-modal image matching between SAR and optical data has become increasingly prominent, and related studies have achieved notable progress. In particular, Deep Learning (DL)-based methods, owing to their strengths in cross-modal feature representation and high-level semantic extraction, have demonstrated excellent matching accuracy and adaptability across varying imaging conditions. However, most publicly available datasets are limited to small image patches and lack complete full-scene image pairs that cover realistic large-scale scenarios, making it difficult to comprehensively evaluate the performance of matching algorithms in practical remote-sensing settings and constraining advances in the training and generalization of DL models. To address these issues, this study develops and releases OSDataset2.0, a large-scale benchmark dataset for SAR-optical image matching. The dataset comprises two parts: A patch-level subset and a scene-level subset. The patch-level subset is composed of 6,476 registered 512 × 512 image pairs covering 14 countries (Argentina, Australia, Poland, Germany, Russia, France, Qatar, Malaysia, the United States, Japan, Türkiye, Singapore, India, and China); the scene-level subset consists of one pair of full-scene optical and SAR images. For full-scene images, high-precision, uniformly distributed ground-truth correspondences are provided, extracted under the principle of imaging-mechanism consistency, together with a general evaluation codebase that supports quantitative analysis of registration accuracy for arbitrary matching algorithms. To further assess the dataset’s effectiveness and challenge level, a systematic evaluation of 11 representative optical-SAR matching methods on OSDataset2.0 is conducted, covering traditional feature-based approaches and mainstream DL models. Experimental results show that the dataset not only supports effective algorithmic comparisons but also provides reliable training resources and a unified evaluation benchmark for subsequent research.
Synthetic Aperture Radar (SAR) and optical imagery are two key remote-sensing modalities in Earth observation, and cross-modal image matching between them is widely applied in tasks such as image fusion, joint interpretation, and high-precision geolocation. In recent years, with the rapid growth of Earth-observation data, the importance of cross-modal image matching between SAR and optical data has become increasingly prominent, and related studies have achieved notable progress. In particular, Deep Learning (DL)-based methods, owing to their strengths in cross-modal feature representation and high-level semantic extraction, have demonstrated excellent matching accuracy and adaptability across varying imaging conditions. However, most publicly available datasets are limited to small image patches and lack complete full-scene image pairs that cover realistic large-scale scenarios, making it difficult to comprehensively evaluate the performance of matching algorithms in practical remote-sensing settings and constraining advances in the training and generalization of DL models. To address these issues, this study develops and releases OSDataset2.0, a large-scale benchmark dataset for SAR-optical image matching. The dataset comprises two parts: A patch-level subset and a scene-level subset. The patch-level subset is composed of 6,476 registered 512 × 512 image pairs covering 14 countries (Argentina, Australia, Poland, Germany, Russia, France, Qatar, Malaysia, the United States, Japan, Türkiye, Singapore, India, and China); the scene-level subset consists of one pair of full-scene optical and SAR images. For full-scene images, high-precision, uniformly distributed ground-truth correspondences are provided, extracted under the principle of imaging-mechanism consistency, together with a general evaluation codebase that supports quantitative analysis of registration accuracy for arbitrary matching algorithms. To further assess the dataset’s effectiveness and challenge level, a systematic evaluation of 11 representative optical-SAR matching methods on OSDataset2.0 is conducted, covering traditional feature-based approaches and mainstream DL models. Experimental results show that the dataset not only supports effective algorithmic comparisons but also provides reliable training resources and a unified evaluation benchmark for subsequent research.
Synthetic Aperture Radar (SAR) offers all-weather, all-day maritime surveillance capabilities. Direct ship detection in the Range Compressed Domain (RCD) eliminates computationally intensive imaging steps—such as range cell migration correction and azimuth compression—thereby considerably improving processing efficiency for near-real-time or real-time applications. However, current detection methods face inherent limitations; traditional constant false alarm rate detectors rely on fixed statistical models and often underperform in complex sea clutter environments. In addition, deep learning approaches heavily rely on annotated data and do not fully leverage phase information; moreover, they exhibit weak interpretability. To address these issues, this paper proposes a self-supervised reinforcement learning framework for ship target detection in the SAR RCD. This framework effectively integrates the physical principles of radar signals with deep reinforcement learning, achieving enhanced detection performance while improving model interpretability and generalization. The framework has the following characteristics: (1) It introduces a reward signal-generation mechanism constrained by statistical scattering models, achieving self-supervised learning without the need for manual annotation; (2) It designs a dual-modal feature-fusion module that can jointly represent amplitude and phase information, effectively retaining the Doppler characteristics of ships; and (3) It adopts a lightweight agent module that integrates a lightweight Q-network, an adaptive feature enhancement module, and a discriminator network; this module reduces computational complexity, meets real-time processing requirements, and enhances the robustness of the model through adversarial training. Experimental results demonstrate that the proposed method achieves an average inference time of only 31.75 s on a large-scale SAR RCD dataset of 20 k×20 k pixels, with a computational load of only 23.81% compared with a two-dimensional convolutional neural network. On a complex-valued RCD dataset, the method attains F1 and recall scores of 50.72% and 54.28%, respectively, outperforming mainstream self-supervised methods by 8.76% and 10.45%, respectively. This study pioneers the application of reinforcement learning to ship detection using SAR RCD, offering a novel approach to robust maritime surveillance by integrating signal modeling and data-driven learning.
Synthetic Aperture Radar (SAR) offers all-weather, all-day maritime surveillance capabilities. Direct ship detection in the Range Compressed Domain (RCD) eliminates computationally intensive imaging steps—such as range cell migration correction and azimuth compression—thereby considerably improving processing efficiency for near-real-time or real-time applications. However, current detection methods face inherent limitations; traditional constant false alarm rate detectors rely on fixed statistical models and often underperform in complex sea clutter environments. In addition, deep learning approaches heavily rely on annotated data and do not fully leverage phase information; moreover, they exhibit weak interpretability. To address these issues, this paper proposes a self-supervised reinforcement learning framework for ship target detection in the SAR RCD. This framework effectively integrates the physical principles of radar signals with deep reinforcement learning, achieving enhanced detection performance while improving model interpretability and generalization. The framework has the following characteristics: (1) It introduces a reward signal-generation mechanism constrained by statistical scattering models, achieving self-supervised learning without the need for manual annotation; (2) It designs a dual-modal feature-fusion module that can jointly represent amplitude and phase information, effectively retaining the Doppler characteristics of ships; and (3) It adopts a lightweight agent module that integrates a lightweight Q-network, an adaptive feature enhancement module, and a discriminator network; this module reduces computational complexity, meets real-time processing requirements, and enhances the robustness of the model through adversarial training. Experimental results demonstrate that the proposed method achieves an average inference time of only 31.75 s on a large-scale SAR RCD dataset of 20 k×20 k pixels, with a computational load of only 23.81% compared with a two-dimensional convolutional neural network. On a complex-valued RCD dataset, the method attains F1 and recall scores of 50.72% and 54.28%, respectively, outperforming mainstream self-supervised methods by 8.76% and 10.45%, respectively. This study pioneers the application of reinforcement learning to ship detection using SAR RCD, offering a novel approach to robust maritime surveillance by integrating signal modeling and data-driven learning.
Specific Emitter Identification (SEI) relies on subtle differences in the radio frequency fingerprints of device-emitted signals to determine the emitter identity attributes. SEI plays a fundamental role in wireless security, spectrum management, and situational awareness. However, as wireless scenarios become increasingly diverse and dynamic, deep learning models trained in a single domain (where the source and target domains share the same distribution) often suffer severe performance degradation in real-world settings such as cross-receiver and cross-time scenarios. This degradation has not yet been comprehensively analyzed. To address this issue, this paper first classifies SEI according to cross-scenario types, and then systematically reviews mainstream algorithm frameworks and representative SEI methods, with a particular focus on the core ideas and key technologies underlying each method. It also summarizes the main open-source cross-scenario SEI datasets. Finally, the paper identifies current research bottlenecks and outlines potential future directions, aiming to facilitate advances in SEI theories and methodologies applicable to complex electromagnetic environments.
Specific Emitter Identification (SEI) relies on subtle differences in the radio frequency fingerprints of device-emitted signals to determine the emitter identity attributes. SEI plays a fundamental role in wireless security, spectrum management, and situational awareness. However, as wireless scenarios become increasingly diverse and dynamic, deep learning models trained in a single domain (where the source and target domains share the same distribution) often suffer severe performance degradation in real-world settings such as cross-receiver and cross-time scenarios. This degradation has not yet been comprehensively analyzed. To address this issue, this paper first classifies SEI according to cross-scenario types, and then systematically reviews mainstream algorithm frameworks and representative SEI methods, with a particular focus on the core ideas and key technologies underlying each method. It also summarizes the main open-source cross-scenario SEI datasets. Finally, the paper identifies current research bottlenecks and outlines potential future directions, aiming to facilitate advances in SEI theories and methodologies applicable to complex electromagnetic environments.
High-speed squint-forward-looking Synthetic Aperture Radar (SAR) imaging (squint angle: >70°) is challenged by severe range-Doppler coupling and Doppler space variance. Traditional Nonlinear Chirp Scaling (NCS) algorithms can effectively mitigate Doppler space variance under high-squint conditions (squint angle: >30°), but they rely on approximate treatments and exhibit rapidly increasing derivation complexity at high scaling orders. This makes high-order generalization difficult and limits their application in high-speed squint-forward-looking SAR systems. To address this issue, this study demonstrates that Fourier Transform (FT) and Inverse FT (IFT) implementations, based on the Principle of Stationary Phase (POSP) and the Method of Series Reversion (MSR) for azimuth data domain transformation, exhibit regular structural patterns. Building on this insight, a fifth-order NCS algorithm with low derivation complexity is proposed, along with a dedicated geometric correction method. For a given predefined slant range model and NCS order, the proposed algorithm requires only a single FT/IFT derivation to obtain the analytical expression of the signal after NCS processing, thereby simplifying both the construction of the Doppler parameter linear equation system and the solution of NCS parameters. This significantly reduces the complexity of the algorithm derivation. Furthermore, an instantaneous projection geometric model is established based on the high-speed squint-forward-looking SAR imaging geometry, enabling the development of a tailored geometric correction method. Compared with traditional NCS algorithms, the proposed fifth-order NCS algorithm achieves superior imaging performance while maintaining computational efficiency. Simulated and real data processing validate its effectiveness and advantages in high-speed squint-forward-looking scenarios.
High-speed squint-forward-looking Synthetic Aperture Radar (SAR) imaging (squint angle: >70°) is challenged by severe range-Doppler coupling and Doppler space variance. Traditional Nonlinear Chirp Scaling (NCS) algorithms can effectively mitigate Doppler space variance under high-squint conditions (squint angle: >30°), but they rely on approximate treatments and exhibit rapidly increasing derivation complexity at high scaling orders. This makes high-order generalization difficult and limits their application in high-speed squint-forward-looking SAR systems. To address this issue, this study demonstrates that Fourier Transform (FT) and Inverse FT (IFT) implementations, based on the Principle of Stationary Phase (POSP) and the Method of Series Reversion (MSR) for azimuth data domain transformation, exhibit regular structural patterns. Building on this insight, a fifth-order NCS algorithm with low derivation complexity is proposed, along with a dedicated geometric correction method. For a given predefined slant range model and NCS order, the proposed algorithm requires only a single FT/IFT derivation to obtain the analytical expression of the signal after NCS processing, thereby simplifying both the construction of the Doppler parameter linear equation system and the solution of NCS parameters. This significantly reduces the complexity of the algorithm derivation. Furthermore, an instantaneous projection geometric model is established based on the high-speed squint-forward-looking SAR imaging geometry, enabling the development of a tailored geometric correction method. Compared with traditional NCS algorithms, the proposed fifth-order NCS algorithm achieves superior imaging performance while maintaining computational efficiency. Simulated and real data processing validate its effectiveness and advantages in high-speed squint-forward-looking scenarios.
This study proposes a processing framework based on Mutual Information Entropy (MIE) and an improved probability hypothesis density filter to address the key challenges—high clutter density and low detection probability—in Passive Bistatic Radar (PBR) target tracking. First, statistical differences in the correlation between target and clutter points, as well as between reference models, are quantified as mutual information entropy values, which are then used to eliminate clutter points. Second, the classical probability hypothesis density filter is improved through dynamic weight compensation, mitigating particle weight degeneration and reducing the deletion of false targets. This approach effectively resolves issues such as track fragmentation and target loss caused by discontinuous measurements with random intervals under low detection probability. The effectiveness of the proposed framework was verified through simulation experiments, and field test data demonstrated that the proposed method achieves good target-tracking performance in practical applications.
This study proposes a processing framework based on Mutual Information Entropy (MIE) and an improved probability hypothesis density filter to address the key challenges—high clutter density and low detection probability—in Passive Bistatic Radar (PBR) target tracking. First, statistical differences in the correlation between target and clutter points, as well as between reference models, are quantified as mutual information entropy values, which are then used to eliminate clutter points. Second, the classical probability hypothesis density filter is improved through dynamic weight compensation, mitigating particle weight degeneration and reducing the deletion of false targets. This approach effectively resolves issues such as track fragmentation and target loss caused by discontinuous measurements with random intervals under low detection probability. The effectiveness of the proposed framework was verified through simulation experiments, and field test data demonstrated that the proposed method achieves good target-tracking performance in practical applications.
Ground Control Points (GCPs) are essential for improving the positioning accuracy of remote sensing imagery. Their spatial distribution and geometric quality directly affect the reliability of orthorectification. GCPs serve as a critical foundation for ensuring the accuracy of multi-source image fusion, change detection, and quantitative inversion. However, traditional corner reflector deployment presents high costs and implementation difficulties, struggling to meet global application demands. Additionally, existing heterogeneous control points (such as optical imagery and laser altimetry data) exhibit significant modal differences relative to Synthetic Aperture Radar (SAR) imagery, which affects their ability to balance accuracy and robustness. To address these challenges, this study proposes an automatic control point extraction method for high-resolution SAR imagery based on multi-source data. Furthermore, a high-precision orthorectification framework is established using control chips. The method leverages the characteristics of widely distributed pole-like artificial features in urban environments: these features exhibit a body-shadow collaborative structure in optical imagery and a cross-shaped strong scattering response in SAR imagery. First, open-source airport runway data are used to correct Google optical imagery, establishing a planar reference framework. Next, initial positioning optimization for stereo SAR images from ascending and descending orbits is achieved by jointly adjusting optical-SAR and stereo SAR image matching points. Finally, road and parking lot vector data are utilized to extract regions of interest, where strong scattering points are identified using a signal-to-clutter ratio detection algorithm. Three-dimensional spatial coordinates of control points are obtained via point target analysis and stereo positioning techniques. After correcting residual planar errors in stereo SAR images using control point coordinates, control chip data for ascending and descending orbit SAR images are generated. Validation experiments using GaoFen-3 SAR images from multiple regions show that the 3D positioning accuracy of control points extracted from spotlight mode stereo SAR imagery reaches the submeter level. Orthorectification of test images using extracted control points and control chips significantly improves positioning accuracy, as verified by corner reflectors and airborne LiDAR point cloud-based ground truth. Positioning errors are 1.78 pixels (spotlight mode), 1.09 pixels (ultrafine stripmap mode), and 0.82 pixels (fine stripmap mode), corresponding to improvements of 47.2%, 49.3%, and 37.4%, respectively, compared to traditional optical reference image matching correction methods. This study introduces crowdsourced information to assist SAR control point extraction and ascending/descending orbit SAR control chip construction, overcoming the accuracy limitations of optical reference image matching correction. The proposed method provides a scalable approach for high-precision positioning and joint processing of high-resolution SAR imagery.
Ground Control Points (GCPs) are essential for improving the positioning accuracy of remote sensing imagery. Their spatial distribution and geometric quality directly affect the reliability of orthorectification. GCPs serve as a critical foundation for ensuring the accuracy of multi-source image fusion, change detection, and quantitative inversion. However, traditional corner reflector deployment presents high costs and implementation difficulties, struggling to meet global application demands. Additionally, existing heterogeneous control points (such as optical imagery and laser altimetry data) exhibit significant modal differences relative to Synthetic Aperture Radar (SAR) imagery, which affects their ability to balance accuracy and robustness. To address these challenges, this study proposes an automatic control point extraction method for high-resolution SAR imagery based on multi-source data. Furthermore, a high-precision orthorectification framework is established using control chips. The method leverages the characteristics of widely distributed pole-like artificial features in urban environments: these features exhibit a body-shadow collaborative structure in optical imagery and a cross-shaped strong scattering response in SAR imagery. First, open-source airport runway data are used to correct Google optical imagery, establishing a planar reference framework. Next, initial positioning optimization for stereo SAR images from ascending and descending orbits is achieved by jointly adjusting optical-SAR and stereo SAR image matching points. Finally, road and parking lot vector data are utilized to extract regions of interest, where strong scattering points are identified using a signal-to-clutter ratio detection algorithm. Three-dimensional spatial coordinates of control points are obtained via point target analysis and stereo positioning techniques. After correcting residual planar errors in stereo SAR images using control point coordinates, control chip data for ascending and descending orbit SAR images are generated. Validation experiments using GaoFen-3 SAR images from multiple regions show that the 3D positioning accuracy of control points extracted from spotlight mode stereo SAR imagery reaches the submeter level. Orthorectification of test images using extracted control points and control chips significantly improves positioning accuracy, as verified by corner reflectors and airborne LiDAR point cloud-based ground truth. Positioning errors are 1.78 pixels (spotlight mode), 1.09 pixels (ultrafine stripmap mode), and 0.82 pixels (fine stripmap mode), corresponding to improvements of 47.2%, 49.3%, and 37.4%, respectively, compared to traditional optical reference image matching correction methods. This study introduces crowdsourced information to assist SAR control point extraction and ascending/descending orbit SAR control chip construction, overcoming the accuracy limitations of optical reference image matching correction. The proposed method provides a scalable approach for high-precision positioning and joint processing of high-resolution SAR imagery.
Adversarial sample generation is a key research direction for uncovering the vulnerabilities of deep neural networks and improving the robustness of Synthetic Aperture Radar Automatic Target Recognition (SAR ATR) systems. This study proposes an optimal adversarial sample generation method for SAR ATR that jointly optimizes misleading effectiveness and fidelity, aiming to resolve the core contradiction between adversarial effectiveness and visual concealment. The generation process is modeled as a joint optimization problem with the goals of balancing “misleading” and “fidelity”. First, an integrated composite transform attack strategy is designed to enhance attack effectiveness, and a joint measurement model is developed that combines the classification accuracy of the target model with the Learned Perceptual Image Patch Similarity (LPIPS) to quantify the two optimization goals. Next, an improved uniformity-guided multiobjective RIME algorithm is proposed. By integrating the Tent chaotic map, hybrid dynamic weighting, and golden sine guidance, the model is efficiently solved, yielding a set of Pareto-optimal solutions that represent various tradeoff degrees. Finally, the YOLOv10 object detection network is employed to identify perturbations in the samples within the solution set, thereby locating the critical points where disturbances occur and enabling the quantification of optimal parameters. Experiments on MSTAR and MiniSAR datasets show that the proposed ensemble compound transform attack method achieves an average target model recognition accuracy of 8.96% across different ensemble models and classification networks, improving the overall misleading effect by an average of 2.25% compared to other methods. Among them, the complex model increases by an average of 5.56%, while the proposed uniformity-guided multiobjective RIME algorithm improves the solution set diversity and convergence speed by over 25% compared with the comparison method. Using this method, the learned perceptual image patch similarity is maintained at 0.407 and the perturbation factor at 0.031, while classification accuracy decreases to 28.81%, demonstrating a tradeoff between misleading effectiveness and visual fidelity. This parameter maintains effective misleading performance under six different defense strategies, demonstrating strong robustness and providing a new approach and quantitative benchmark for adversarial attack research in SAR ATR.
Adversarial sample generation is a key research direction for uncovering the vulnerabilities of deep neural networks and improving the robustness of Synthetic Aperture Radar Automatic Target Recognition (SAR ATR) systems. This study proposes an optimal adversarial sample generation method for SAR ATR that jointly optimizes misleading effectiveness and fidelity, aiming to resolve the core contradiction between adversarial effectiveness and visual concealment. The generation process is modeled as a joint optimization problem with the goals of balancing “misleading” and “fidelity”. First, an integrated composite transform attack strategy is designed to enhance attack effectiveness, and a joint measurement model is developed that combines the classification accuracy of the target model with the Learned Perceptual Image Patch Similarity (LPIPS) to quantify the two optimization goals. Next, an improved uniformity-guided multiobjective RIME algorithm is proposed. By integrating the Tent chaotic map, hybrid dynamic weighting, and golden sine guidance, the model is efficiently solved, yielding a set of Pareto-optimal solutions that represent various tradeoff degrees. Finally, the YOLOv10 object detection network is employed to identify perturbations in the samples within the solution set, thereby locating the critical points where disturbances occur and enabling the quantification of optimal parameters. Experiments on MSTAR and MiniSAR datasets show that the proposed ensemble compound transform attack method achieves an average target model recognition accuracy of 8.96% across different ensemble models and classification networks, improving the overall misleading effect by an average of 2.25% compared to other methods. Among them, the complex model increases by an average of 5.56%, while the proposed uniformity-guided multiobjective RIME algorithm improves the solution set diversity and convergence speed by over 25% compared with the comparison method. Using this method, the learned perceptual image patch similarity is maintained at 0.407 and the perturbation factor at 0.031, while classification accuracy decreases to 28.81%, demonstrating a tradeoff between misleading effectiveness and visual fidelity. This parameter maintains effective misleading performance under six different defense strategies, demonstrating strong robustness and providing a new approach and quantitative benchmark for adversarial attack research in SAR ATR.
Passive localization methods based on synthetic aperture imaging offer high positioning accuracy. However, in scenarios involving multiple radar emitters transmitting Linear Frequency-Modulated (LFM) signals, distinguishing signals that are overlapped in the time and frequency domains can be challenging. This phenomenon, known as phase overlap, results in a significant degradation of localization performance. To address this issue, the present paper proposes a single-satellite multi-radar-emitter passive localization method based on synthetic aperture imaging using time-frequency parameter estimation. First, a signal model for multiple radar emitters transmitting LFM signals is constructed. The time-frequency parameters of the multiple radar emitter signals are estimated concurrently via a combination of Short-Time Fourier Transform (STFT) and DBSCAN. A rapid approximation of the azimuth chirp rate is attained through a coarse-to-fine search strategy founded upon the use of the STFT. The accurate localization of multiple radar emitters is ultimately realized through the implementation of two-dimensional focusing in the range and azimuth dimensions. The Cramer-Rao lower bound of the proposed method is derived on this basis. The experimental findings demonstrate that the proposed method enhances the localization accuracy by approximately 10 km at a signal-to-noise ratio of −10 dB, in comparison with the enhanced real-valued space-time subspace data fusion-based direct positioning method. Moreover, it reduces the computational time by half relative to the CLEAN-based synthetic aperture multi-source localization approach.
Passive localization methods based on synthetic aperture imaging offer high positioning accuracy. However, in scenarios involving multiple radar emitters transmitting Linear Frequency-Modulated (LFM) signals, distinguishing signals that are overlapped in the time and frequency domains can be challenging. This phenomenon, known as phase overlap, results in a significant degradation of localization performance. To address this issue, the present paper proposes a single-satellite multi-radar-emitter passive localization method based on synthetic aperture imaging using time-frequency parameter estimation. First, a signal model for multiple radar emitters transmitting LFM signals is constructed. The time-frequency parameters of the multiple radar emitter signals are estimated concurrently via a combination of Short-Time Fourier Transform (STFT) and DBSCAN. A rapid approximation of the azimuth chirp rate is attained through a coarse-to-fine search strategy founded upon the use of the STFT. The accurate localization of multiple radar emitters is ultimately realized through the implementation of two-dimensional focusing in the range and azimuth dimensions. The Cramer-Rao lower bound of the proposed method is derived on this basis. The experimental findings demonstrate that the proposed method enhances the localization accuracy by approximately 10 km at a signal-to-noise ratio of −10 dB, in comparison with the enhanced real-valued space-time subspace data fusion-based direct positioning method. Moreover, it reduces the computational time by half relative to the CLEAN-based synthetic aperture multi-source localization approach.
This study addresses the critical challenge of radar target tracking in complex maritime environments. Although conventional feature-aided Bayesian tracking methods have advanced in maritime radar applications, their robustness considerably deteriorates in scenarios with severe sea clutter and interacting targets. To overcome these limitations, an Adaptive Multifocus Correlation Filter with Bayesian Fusion (AMFCF-BF) is proposed herein. The method constructs multiple subviews within the probabilistic distribution of the target state, with each subview assigned an independent correlation filter to generate a local response map, enabling multihypothesis state modeling. During iterative tracking, these response maps are used to estimate states and dynamically guide the focusing of subviews toward high-confidence regions, enhancing adaptability to complex target maneuvers. To further mitigate false alarms and missed detections caused by strong sea clutter, a virtual-view simulation based focusing model is developed, which effectively suppresses filter drift under adverse conditions. Finally, all subview responses are fused within a Bayesian multimeasurement framework to produce a globally consistent target-state estimate. Experimental results using simulated and real maritime radar data demonstrate that the proposed AMFCF-BF achieves an average center location error of 3.47 pixels, reducing tracking error by ～70% compared with typical feature-assisted correlation filtering methods. In terms of location precision, the proposed filter achieves an overall improvement of ～21%, showing significantly enhanced tracking accuracy and anti-interference performance, validating the effectiveness of the multifocus correlation filtering mechanism and Bayesian fusion strategy.
This study addresses the critical challenge of radar target tracking in complex maritime environments. Although conventional feature-aided Bayesian tracking methods have advanced in maritime radar applications, their robustness considerably deteriorates in scenarios with severe sea clutter and interacting targets. To overcome these limitations, an Adaptive Multifocus Correlation Filter with Bayesian Fusion (AMFCF-BF) is proposed herein. The method constructs multiple subviews within the probabilistic distribution of the target state, with each subview assigned an independent correlation filter to generate a local response map, enabling multihypothesis state modeling. During iterative tracking, these response maps are used to estimate states and dynamically guide the focusing of subviews toward high-confidence regions, enhancing adaptability to complex target maneuvers. To further mitigate false alarms and missed detections caused by strong sea clutter, a virtual-view simulation based focusing model is developed, which effectively suppresses filter drift under adverse conditions. Finally, all subview responses are fused within a Bayesian multimeasurement framework to produce a globally consistent target-state estimate. Experimental results using simulated and real maritime radar data demonstrate that the proposed AMFCF-BF achieves an average center location error of 3.47 pixels, reducing tracking error by ～70% compared with typical feature-assisted correlation filtering methods. In terms of location precision, the proposed filter achieves an overall improvement of ～21%, showing significantly enhanced tracking accuracy and anti-interference performance, validating the effectiveness of the multifocus correlation filtering mechanism and Bayesian fusion strategy.
Vortex Electromagnetic Wave Radar (VEWR) leverages the orthogonality of Orbital Angular Momentum (OAM) modes, introducing a new physical dimension that theoretically overcomes the azimuth resolution limitations of conventional radar systems and enables enhanced micro-motion perception and forward-looking imaging. However, in practical engineering applications, the limited number of available OAM modes and the presence of complex electromagnetic noise often cause severe mode aliasing and resolution degradation. Existing sparse imaging methods face inherent trade-offs between accuracy and computational efficiency and exhibit limited robustness to noise. To address these issues, this paper proposes a super-resolution imaging framework that integrates Mode Correlation Weighting and Adaptive Regularization (MCWAR). First, a forward-looking imaging geometry and a wavefront-modulated signal model for VEWR are established. Subsequently, an OAM mode correlation matrix is designed to characterize the nonuniform distribution of radiation energy among modes, where Bessel-function-modulated weights reinforce the low-rank constraints of dominant radiation components. Finally, a compound optimization model combining sparsity and low-rankness priors is developed, incorporating an adaptive weighting mechanism that dynamically balances structural preservation and noise suppression. A joint optimization framework based on the Alternating Direction Method of Multipliers (ADMM) and Augmented Lagrange Multiplier (ALM) algorithms is constructed, in which the core image-updating subproblem is efficiently solved using a momentum-accelerated Two-Dimensional Conjugate Gradient Least Squares (2D-CGLS) method. Numerical simulations and electromagnetic experiments verify that the proposed method preserves target structural integrity under limited modes and strong noise, while effectively improving both computational efficiency and imaging quality.
Vortex Electromagnetic Wave Radar (VEWR) leverages the orthogonality of Orbital Angular Momentum (OAM) modes, introducing a new physical dimension that theoretically overcomes the azimuth resolution limitations of conventional radar systems and enables enhanced micro-motion perception and forward-looking imaging. However, in practical engineering applications, the limited number of available OAM modes and the presence of complex electromagnetic noise often cause severe mode aliasing and resolution degradation. Existing sparse imaging methods face inherent trade-offs between accuracy and computational efficiency and exhibit limited robustness to noise. To address these issues, this paper proposes a super-resolution imaging framework that integrates Mode Correlation Weighting and Adaptive Regularization (MCWAR). First, a forward-looking imaging geometry and a wavefront-modulated signal model for VEWR are established. Subsequently, an OAM mode correlation matrix is designed to characterize the nonuniform distribution of radiation energy among modes, where Bessel-function-modulated weights reinforce the low-rank constraints of dominant radiation components. Finally, a compound optimization model combining sparsity and low-rankness priors is developed, incorporating an adaptive weighting mechanism that dynamically balances structural preservation and noise suppression. A joint optimization framework based on the Alternating Direction Method of Multipliers (ADMM) and Augmented Lagrange Multiplier (ALM) algorithms is constructed, in which the core image-updating subproblem is efficiently solved using a momentum-accelerated Two-Dimensional Conjugate Gradient Least Squares (2D-CGLS) method. Numerical simulations and electromagnetic experiments verify that the proposed method preserves target structural integrity under limited modes and strong noise, while effectively improving both computational efficiency and imaging quality.
High-Resolution Wide-Swath (HRWS) imaging is a key development direction for next-generation spaceborne Synthetic Aperture Radar (SAR) systems. Multiple-Input Multiple-Output (MIMO) SAR systems offer high spatial degrees of freedom, enabling enhanced system performance. However, effectively separating echoes from different transmit channels in MIMO-SAR systems is key to unlocking their advantages in spatial degrees of freedom. In this regard, a novel Space-Time Phase-Coded (STPC) waveform for MIMO-SAR systems is proposed based on the phase characteristics of SAR signals and the space-time properties of the “stop-and-go” model. This waveform modulates transmitted signals in the range dimension via phase coding and emits them at distinct spatial positions within each pulse repetition period, following a preset coding sequence. Upon reception, demodulating aliased echoes using receiver timing matched to the transmitter enables the efficient separation of echoes from different transmit channels. The proposed scheme can be integrated with existing classical azimuth multichannel reconstruction methods, effectively mitigating the trade-off between Pulse Repetition Frequency (PRF) and echo separability. Compared with the Alamouti, Short-Term Shift-Orthogonal (STSO), and Segmented Phase Code (SPC) waveforms in current MIMO-SAR systems, the STPC approach reduces antenna requirements by nearly 50%, thereby lowering the cost and complexity of hardware implementation. Simulation experiments on point targets and distributed scenes verify that the proposed waveform and processing scheme effectively suppress interwaveform interference and deliver strong imaging performance.
High-Resolution Wide-Swath (HRWS) imaging is a key development direction for next-generation spaceborne Synthetic Aperture Radar (SAR) systems. Multiple-Input Multiple-Output (MIMO) SAR systems offer high spatial degrees of freedom, enabling enhanced system performance. However, effectively separating echoes from different transmit channels in MIMO-SAR systems is key to unlocking their advantages in spatial degrees of freedom. In this regard, a novel Space-Time Phase-Coded (STPC) waveform for MIMO-SAR systems is proposed based on the phase characteristics of SAR signals and the space-time properties of the “stop-and-go” model. This waveform modulates transmitted signals in the range dimension via phase coding and emits them at distinct spatial positions within each pulse repetition period, following a preset coding sequence. Upon reception, demodulating aliased echoes using receiver timing matched to the transmitter enables the efficient separation of echoes from different transmit channels. The proposed scheme can be integrated with existing classical azimuth multichannel reconstruction methods, effectively mitigating the trade-off between Pulse Repetition Frequency (PRF) and echo separability. Compared with the Alamouti, Short-Term Shift-Orthogonal (STSO), and Segmented Phase Code (SPC) waveforms in current MIMO-SAR systems, the STPC approach reduces antenna requirements by nearly 50%, thereby lowering the cost and complexity of hardware implementation. Simulation experiments on point targets and distributed scenes verify that the proposed waveform and processing scheme effectively suppress interwaveform interference and deliver strong imaging performance.
To enhance the jamming recognition capabilities of radars in complex electromagnetic environments, this study proposes YOLO-S3, a lightweight network for recognizing composite jamming signals. YOLO-S3 is characterized by three core attributes: Smartness, slimness, and high speed. Initially, a technical approach based on visual detection algorithms is introduced to identify 2D time-frequency representations of jamming signals. An image dataset of composite jamming signals is constructed using signal modeling, simulation technology, and the short-time Fourier transform. Next, the backbone and neck networks of YOLOv8n are restructured by integrating StarNet and SlimNeck, and a Self-Attention Detect Head (SADH) is designed to enhance feature extraction. These modifications result in a lightweight network without compromising recognition accuracy. Finally, the network’s performance is validated through ablation and comparative experiments. Results show that YOLO-S3 features a highly lightweight network design. When the signal-to-jamming ratio varies from −10 to 0 dB and the Signal-to-Noise Ratio (SNR) is ≥0 dB, the network achieves an impressive average recognition accuracy of 99.5%. Even when the SNR decreases to −10 dB, it maintains a robust average recognition accuracy of 95.5%, exhibiting strong performance under low SNR conditions. These findings provide a promising solution for the real-time recognition of composite jamming signals on resource-constrained platforms such as airborne radar signal processors and portable electronic devices.
To enhance the jamming recognition capabilities of radars in complex electromagnetic environments, this study proposes YOLO-S3, a lightweight network for recognizing composite jamming signals. YOLO-S3 is characterized by three core attributes: Smartness, slimness, and high speed. Initially, a technical approach based on visual detection algorithms is introduced to identify 2D time-frequency representations of jamming signals. An image dataset of composite jamming signals is constructed using signal modeling, simulation technology, and the short-time Fourier transform. Next, the backbone and neck networks of YOLOv8n are restructured by integrating StarNet and SlimNeck, and a Self-Attention Detect Head (SADH) is designed to enhance feature extraction. These modifications result in a lightweight network without compromising recognition accuracy. Finally, the network’s performance is validated through ablation and comparative experiments. Results show that YOLO-S3 features a highly lightweight network design. When the signal-to-jamming ratio varies from −10 to 0 dB and the Signal-to-Noise Ratio (SNR) is ≥0 dB, the network achieves an impressive average recognition accuracy of 99.5%. Even when the SNR decreases to −10 dB, it maintains a robust average recognition accuracy of 95.5%, exhibiting strong performance under low SNR conditions. These findings provide a promising solution for the real-time recognition of composite jamming signals on resource-constrained platforms such as airborne radar signal processors and portable electronic devices.
Non-cooperative bistatic radar exhibits significant application value for both civilian and military applications due to its anti-stealth and anti-jamming capabilities. However, its practical implementation faces challenges from unavoidable multipath interference and noise contamination in reference signals, stemming from uncontrollable radar illuminators and complex geographical environments. These effects substantially degrade the performance of cross-correlation processing between reference and echo signals compared with the ideal matched filter, resulting in stationary false targets. Such issues remain a critical bottleneck to operational deployment. This study systematically addresses these challenges by analyzing cross-correlation degradation under multipath and noise in the reference channel, and by establishing a quantitative mapping between multipath intensity, noise power, and detection probability. For Linear Frequency Modulated (LFM) signals, a dechirp-based multipath suppression algorithm is proposed. The algorithm exploits the inherent properties of LFM signals, transforming multipath components with different delays into distinct frequency offsets. Compared with mainstream Fractional Fourier Transform (FrFT) methods, this approach exhibits greater frequency separation among multipath components, enabling effective suppression with significantly reduced filter orders. The algorithm outperforms conventional methods in improving overall detection probability. Measured data processing in practical field-test scenarios (direct-path signals overwhelmed by strong multipath interference) validates the method’s efficacy in eliminating false targets, correcting range offsets, and enhancing detection probability.
Non-cooperative bistatic radar exhibits significant application value for both civilian and military applications due to its anti-stealth and anti-jamming capabilities. However, its practical implementation faces challenges from unavoidable multipath interference and noise contamination in reference signals, stemming from uncontrollable radar illuminators and complex geographical environments. These effects substantially degrade the performance of cross-correlation processing between reference and echo signals compared with the ideal matched filter, resulting in stationary false targets. Such issues remain a critical bottleneck to operational deployment. This study systematically addresses these challenges by analyzing cross-correlation degradation under multipath and noise in the reference channel, and by establishing a quantitative mapping between multipath intensity, noise power, and detection probability. For Linear Frequency Modulated (LFM) signals, a dechirp-based multipath suppression algorithm is proposed. The algorithm exploits the inherent properties of LFM signals, transforming multipath components with different delays into distinct frequency offsets. Compared with mainstream Fractional Fourier Transform (FrFT) methods, this approach exhibits greater frequency separation among multipath components, enabling effective suppression with significantly reduced filter orders. The algorithm outperforms conventional methods in improving overall detection probability. Measured data processing in practical field-test scenarios (direct-path signals overwhelmed by strong multipath interference) validates the method’s efficacy in eliminating false targets, correcting range offsets, and enhancing detection probability.
Low-altitude targets, represented by rotor unmanned aerial vehicles, can typically adopt a slow-cruise mode. As a result, their echoes fall within the Doppler Blind Zone (DBZ) and evade radar detection and tracking. The cluttered low-altitude environment adds to further complexity. To address this issue, this study proposes a method grounded in the framework of random finite set and designed for tracking slow-moving targets with a low-altitude surveillance radar. Inspired by the Bayesian occupancy filter, the proposed method initially models the radar Field of View (FoV) as a grid map. It is uniformly partitioned along the angle-range axis, ensuring that each cell captures a specific segment of the FoV. Then, adaptive filtering parameter modules are meticulously designed by leveraging the distinct dynamic characteristics of slow-moving targets and ground clutter. Subsequently, a probability hypothesis density filter is deployed to conduct unified filtering on the grid map situated within the DBZ. The final step involves the use of clustering methods to extract information about the target of interest. Simulation results validate the effectiveness, robustness, and superior performance of the proposed method across typical surveillance scenarios involving multiple slow-moving targets, noise, and clutter.
Low-altitude targets, represented by rotor unmanned aerial vehicles, can typically adopt a slow-cruise mode. As a result, their echoes fall within the Doppler Blind Zone (DBZ) and evade radar detection and tracking. The cluttered low-altitude environment adds to further complexity. To address this issue, this study proposes a method grounded in the framework of random finite set and designed for tracking slow-moving targets with a low-altitude surveillance radar. Inspired by the Bayesian occupancy filter, the proposed method initially models the radar Field of View (FoV) as a grid map. It is uniformly partitioned along the angle-range axis, ensuring that each cell captures a specific segment of the FoV. Then, adaptive filtering parameter modules are meticulously designed by leveraging the distinct dynamic characteristics of slow-moving targets and ground clutter. Subsequently, a probability hypothesis density filter is deployed to conduct unified filtering on the grid map situated within the DBZ. The final step involves the use of clustering methods to extract information about the target of interest. Simulation results validate the effectiveness, robustness, and superior performance of the proposed method across typical surveillance scenarios involving multiple slow-moving targets, noise, and clutter.
To address interruptions in phantom tracks caused by platform failures or damage during Unmanned Aerial Vehicle (UAV) swarm deception operations against radar networks, this study proposes a game theory-based joint optimization algorithm for UAV swarm task allocation and trajectory planning. A decentralized swarm cooperation mechanism is designed to create a cooperative game model for UAV phantom track deception against radar networks. Based on the radar network homology test criterion, an optimization model is developed to maximize the utility function of the phantom track deception game, subject to constraints on UAV swarm kinematic performance and task allocation requirements. The existence and convergence of a Nash equilibrium are rigorously proven using exact potential game theory. To address the resulting non-convex, non-linear, mixed-integer optimization problem, an iterative algorithm is developed that combines distributed coalition game theory with a genetic algorithm. The simulation results demonstrate that, compared with existing approaches, the proposed algorithm effectively replans deception tasks and trajectories in response to platform failures or damage, thereby enhancing the continuity and effectiveness of phantom track generation against radar networks.
To address interruptions in phantom tracks caused by platform failures or damage during Unmanned Aerial Vehicle (UAV) swarm deception operations against radar networks, this study proposes a game theory-based joint optimization algorithm for UAV swarm task allocation and trajectory planning. A decentralized swarm cooperation mechanism is designed to create a cooperative game model for UAV phantom track deception against radar networks. Based on the radar network homology test criterion, an optimization model is developed to maximize the utility function of the phantom track deception game, subject to constraints on UAV swarm kinematic performance and task allocation requirements. The existence and convergence of a Nash equilibrium are rigorously proven using exact potential game theory. To address the resulting non-convex, non-linear, mixed-integer optimization problem, an iterative algorithm is developed that combines distributed coalition game theory with a genetic algorithm. The simulation results demonstrate that, compared with existing approaches, the proposed algorithm effectively replans deception tasks and trajectories in response to platform failures or damage, thereby enhancing the continuity and effectiveness of phantom track generation against radar networks.
The effective utilization of Synthetic Aperture Radar (SAR) adversarial examples enables specific targets to achieve remote sensing stealth against intelligent detection systems, thereby evading detection and recognition by adversaries. Digital domain SAR adversarial methods, which operate exclusively in the image domain, produce adversarial images that are not physically realizable and therefore cannot generated by real SAR imaging systems. Existing physical domain approaches typically involve deploying corner reflectors or electromagnetic metasurfaces around targets and simulating adversarial examples using via computational electromagnetics. However, the limited accuracy of scattering estimation often constrains the practical protective efficacy of these methods. To overcome these limitations, this paper proposes an active jammer-based adversarial attack method that integrates SAR active jamming technology with adversarial attack methods to generate adversarial examples by perturbing the target’s echo signals in the signal domain. First, a multiple-phase sectionalized modulation jamming method based on cosine amplitude weighting is selected, enabling parameterized control of the adversarial jamming signal through the design of perturbation components. Next, the adversarial jamming signal generated by the active jammer is fused with the target’s echo signal according to the principles and actual processes of SAR imaging and is then subjected to imaging processing to produce physically realizable SAR adversarial examples. Finally, the differential evolution algorithm is employed to dynamically adjust parameters, such as the energy distribution and jamming range of the adversarial jamming signal, thereby optimizing the SAR adversarial examples to achieve optimal attack success rates even with minimal interference intensity. Experimental results on the MSTAR dataset, a widely used benchmark in the field of SAR Automatic Target Recognition (ATR), show that the proposed method achieves an average fooling rate of 90.88% and demonstrates superior transferability across five different SAR ATR models, with the highest transfer fooling rate reaching 75.57%. Overall, the proposed method generates more physically realizable adversarial examples compared with existing digital domain methods, effectively protecting specific targets in remote sensing detection and providing guidance for the practical application of active jamming signals in real-world scenarios.
The effective utilization of Synthetic Aperture Radar (SAR) adversarial examples enables specific targets to achieve remote sensing stealth against intelligent detection systems, thereby evading detection and recognition by adversaries. Digital domain SAR adversarial methods, which operate exclusively in the image domain, produce adversarial images that are not physically realizable and therefore cannot generated by real SAR imaging systems. Existing physical domain approaches typically involve deploying corner reflectors or electromagnetic metasurfaces around targets and simulating adversarial examples using via computational electromagnetics. However, the limited accuracy of scattering estimation often constrains the practical protective efficacy of these methods. To overcome these limitations, this paper proposes an active jammer-based adversarial attack method that integrates SAR active jamming technology with adversarial attack methods to generate adversarial examples by perturbing the target’s echo signals in the signal domain. First, a multiple-phase sectionalized modulation jamming method based on cosine amplitude weighting is selected, enabling parameterized control of the adversarial jamming signal through the design of perturbation components. Next, the adversarial jamming signal generated by the active jammer is fused with the target’s echo signal according to the principles and actual processes of SAR imaging and is then subjected to imaging processing to produce physically realizable SAR adversarial examples. Finally, the differential evolution algorithm is employed to dynamically adjust parameters, such as the energy distribution and jamming range of the adversarial jamming signal, thereby optimizing the SAR adversarial examples to achieve optimal attack success rates even with minimal interference intensity. Experimental results on the MSTAR dataset, a widely used benchmark in the field of SAR Automatic Target Recognition (ATR), show that the proposed method achieves an average fooling rate of 90.88% and demonstrates superior transferability across five different SAR ATR models, with the highest transfer fooling rate reaching 75.57%. Overall, the proposed method generates more physically realizable adversarial examples compared with existing digital domain methods, effectively protecting specific targets in remote sensing detection and providing guidance for the practical application of active jamming signals in real-world scenarios.
Salt lakes, rich in potassium and lithium mineral resources, are typically mined using the salt field crystallization method. Specifically, brine is first moved to sodium salt fields where sodium salts crystallize, and then it is moved to potassium salt fields for the precipitation of potassium salts. Determining the type of salt field is essential for accurately estimating salt production and ensuring efficient mining operations. Because different types of salt fields exhibit different salt precipitation rates, they also produce distinct variations in scattering intensity that can be observed in multi-temporal Polarimetric Synthetic Aperture Radar (PolSAR) data. To explore this property, this study proposes a salt field classification method based on multi-temporal PolSAR. First, to accurately characterize the long-term scattering variations in salt fields, a new multi-temporal polarization feature, i.e., dominant scattering temporal entropy, is introduced. The main scattering mechanism of the target area is extracted from the polarimetric covariance matrix, from which the temporal correlation between any two PolSAR images is calculated to construct a temporal correlation matrix. The principal change direction and magnitude of scattering variation in land cover across the time series are then obtained from the temporal correlation matrix through diagonalization, and entropy is used to quantify change intensity and provide an accurate measure of cumulative change. Second, this study demonstrates that the dominant scattering temporal entropy follows the Gaussian distribution, enabling the design of a classifier based on Chernoff distance. Classification is performed by comparing the Chernoff distance of entropy probability distributions within superpixels. The proposed method achieves overall classification accuracies of 84.13% and 86.13% on the Qarhan Salt Lake and Dead Sea Sentinel-1 datasets, respectively, representing an improvement of about 10% over existing time-series PolSAR methods. The classification results exhibit superior spatial consistency and noise robustness compared with other methods.
Salt lakes, rich in potassium and lithium mineral resources, are typically mined using the salt field crystallization method. Specifically, brine is first moved to sodium salt fields where sodium salts crystallize, and then it is moved to potassium salt fields for the precipitation of potassium salts. Determining the type of salt field is essential for accurately estimating salt production and ensuring efficient mining operations. Because different types of salt fields exhibit different salt precipitation rates, they also produce distinct variations in scattering intensity that can be observed in multi-temporal Polarimetric Synthetic Aperture Radar (PolSAR) data. To explore this property, this study proposes a salt field classification method based on multi-temporal PolSAR. First, to accurately characterize the long-term scattering variations in salt fields, a new multi-temporal polarization feature, i.e., dominant scattering temporal entropy, is introduced. The main scattering mechanism of the target area is extracted from the polarimetric covariance matrix, from which the temporal correlation between any two PolSAR images is calculated to construct a temporal correlation matrix. The principal change direction and magnitude of scattering variation in land cover across the time series are then obtained from the temporal correlation matrix through diagonalization, and entropy is used to quantify change intensity and provide an accurate measure of cumulative change. Second, this study demonstrates that the dominant scattering temporal entropy follows the Gaussian distribution, enabling the design of a classifier based on Chernoff distance. Classification is performed by comparing the Chernoff distance of entropy probability distributions within superpixels. The proposed method achieves overall classification accuracies of 84.13% and 86.13% on the Qarhan Salt Lake and Dead Sea Sentinel-1 datasets, respectively, representing an improvement of about 10% over existing time-series PolSAR methods. The classification results exhibit superior spatial consistency and noise robustness compared with other methods.
In recent years, bionic super-resolution technology, inspired by biological perception mechanisms, has emerged as a substantial research direction aimed at overcoming the limitations of radar resolution. The Baseband Spectrogram Correlation and Transformation (BSCT) model, which is based on bat hearing, offers a novel approach to enhancing traditional radar resolution. However, the model exhibits inherent limitations, including insufficient multi-target adaptability and the inability to utilize polarization information. To address these problems, this paper proposes a polarization-enhanced bionic super-resolution model: Polarimetric Baseband Spectrogram Correlation and Transformation (P-BSCT) for Mechanical Rotation Polarimetric Radar (MRPR). The primary contributions of this study are as follows: first, the integration of the bat BSCT model with MRPR, thereby enabling the utilization of polarization information and the execution of polarization measurements; second, the proposal of an advanced signal processing method, which overcomes the limitations of the original BSCT in two-target and static scenes, effectively applying to multi-target and moving-target scenarios, and exhibiting no impact on the resolution effect due to signal modulation. P-BSCT has been demonstrated to enhance resolving power by approximately 15 dB under optimal conditions when compared with the original BSCT model. In scenarios involving moving targets, targets exhibiting equivalent polarization scattering properties, and nonlinear FM signals, the resolving performance of P-BSCT remains essentially unchanged, demonstrating notable robustness.
In recent years, bionic super-resolution technology, inspired by biological perception mechanisms, has emerged as a substantial research direction aimed at overcoming the limitations of radar resolution. The Baseband Spectrogram Correlation and Transformation (BSCT) model, which is based on bat hearing, offers a novel approach to enhancing traditional radar resolution. However, the model exhibits inherent limitations, including insufficient multi-target adaptability and the inability to utilize polarization information. To address these problems, this paper proposes a polarization-enhanced bionic super-resolution model: Polarimetric Baseband Spectrogram Correlation and Transformation (P-BSCT) for Mechanical Rotation Polarimetric Radar (MRPR). The primary contributions of this study are as follows: first, the integration of the bat BSCT model with MRPR, thereby enabling the utilization of polarization information and the execution of polarization measurements; second, the proposal of an advanced signal processing method, which overcomes the limitations of the original BSCT in two-target and static scenes, effectively applying to multi-target and moving-target scenarios, and exhibiting no impact on the resolution effect due to signal modulation. P-BSCT has been demonstrated to enhance resolving power by approximately 15 dB under optimal conditions when compared with the original BSCT model. In scenarios involving moving targets, targets exhibiting equivalent polarization scattering properties, and nonlinear FM signals, the resolving performance of P-BSCT remains essentially unchanged, demonstrating notable robustness.
A single Synthetic Aperture Radar (SAR) image can capture only two-dimensional information, and traditional multitemporal Interferometric SAR (InSAR) techniques struggle with the layover problem, particularly in urban areas. SAR Tomography (TomoSAR) provides the advantage of obtaining three-dimensional (3-D) information while offering a feasible solution to the layover problem. This technique relies on repeated observations of the target scene to achieve 3-D resolution by synthesizing the aperture in the elevation direction. In China, early data sources for spaceborne TomoSAR primarily came from foreign satellites such as TerraSAR-X and COSMO-SkyMed, which constrained the development of the country’s TomoSAR technology. In recent years, the launch of Chinese commercial SAR satellites (e.g., Fucheng-1 and Hongtu-1) has expanded the range of data acquisition sources. However, studies on the tomographic 3-D inversion of urban buildings and structures using data from Chinese commercial SAR satellites remain limited. To validate the usability of Chinese commercial SAR satellite data in urban tomography 3-D parameter inversion and the effectiveness of applying these data to existing tomography imaging methods, this paper develops a 3-D inversion framework for urban TomoSAR and conducts a 3-D inversion study of urban buildings and structures using data from the Fucheng-1 satellite of Spacety Co., Ltd. (Changsha) and the Hongtu-1 SAR satellite of Piesat Information Technology Co., Ltd. The experimental results validate the potential of these two satellite systems for tomographic applications, providing pioneering technical support for future in-depth research and applications.
A single Synthetic Aperture Radar (SAR) image can capture only two-dimensional information, and traditional multitemporal Interferometric SAR (InSAR) techniques struggle with the layover problem, particularly in urban areas. SAR Tomography (TomoSAR) provides the advantage of obtaining three-dimensional (3-D) information while offering a feasible solution to the layover problem. This technique relies on repeated observations of the target scene to achieve 3-D resolution by synthesizing the aperture in the elevation direction. In China, early data sources for spaceborne TomoSAR primarily came from foreign satellites such as TerraSAR-X and COSMO-SkyMed, which constrained the development of the country’s TomoSAR technology. In recent years, the launch of Chinese commercial SAR satellites (e.g., Fucheng-1 and Hongtu-1) has expanded the range of data acquisition sources. However, studies on the tomographic 3-D inversion of urban buildings and structures using data from Chinese commercial SAR satellites remain limited. To validate the usability of Chinese commercial SAR satellite data in urban tomography 3-D parameter inversion and the effectiveness of applying these data to existing tomography imaging methods, this paper develops a 3-D inversion framework for urban TomoSAR and conducts a 3-D inversion study of urban buildings and structures using data from the Fucheng-1 satellite of Spacety Co., Ltd. (Changsha) and the Hongtu-1 SAR satellite of Piesat Information Technology Co., Ltd. The experimental results validate the potential of these two satellite systems for tomographic applications, providing pioneering technical support for future in-depth research and applications.
Low-sidelobe waveforms are fundamental for ensuring the basic detection performance of radars. Designing waveforms with low sidelobes in the range dimension, the velocity dimension, or both, remains a major challenge in radar research. To address the issue of sidelobe suppression in the velocity dimension for coherent pulse trains, this paper proposes a joint design method of variable-pulse-width pulse trains and receive mismatched filtering. The proposed method uses a symmetric positive window function to directly construct both the pulse width sequence and the receive weighting sequence. As a result, the characteristics of the window function’s amplitude spectrum, including low sidelobes and the broadening of the 3-dB mainlobe, are transferred into the mismatched filtering output of the coherent pulse train. Theoretical analysis shows that the proposed method incurs a smaller mismatched Signal-to-Noise Ratio (SNR) loss than when the window function is applied solely for receive mismatched filtering. The effects of window functions and the minimum pulse-width constraint on SNR loss and weak target detection performance under strong target interference are analyzed through simulations, illustrating the advantages of the proposed joint transceiver design method.
Low-sidelobe waveforms are fundamental for ensuring the basic detection performance of radars. Designing waveforms with low sidelobes in the range dimension, the velocity dimension, or both, remains a major challenge in radar research. To address the issue of sidelobe suppression in the velocity dimension for coherent pulse trains, this paper proposes a joint design method of variable-pulse-width pulse trains and receive mismatched filtering. The proposed method uses a symmetric positive window function to directly construct both the pulse width sequence and the receive weighting sequence. As a result, the characteristics of the window function’s amplitude spectrum, including low sidelobes and the broadening of the 3-dB mainlobe, are transferred into the mismatched filtering output of the coherent pulse train. Theoretical analysis shows that the proposed method incurs a smaller mismatched Signal-to-Noise Ratio (SNR) loss than when the window function is applied solely for receive mismatched filtering. The effects of window functions and the minimum pulse-width constraint on SNR loss and weak target detection performance under strong target interference are analyzed through simulations, illustrating the advantages of the proposed joint transceiver design method.
微信 | 公众平台 
