Submit Manuscript
Peer Review
Editor WorkMost Downloaded
1
2026, 15(2): 387-408.
Abstract
4613KB
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.
2
2023, 12(2): 456-469.
Marine target detection and recognition depend on the characteristics of marine targets and sea clutter. Therefore, understanding the essential features of marine targets based on the measured data is crucial for advancing target detection and recognition technology. To address the issue of insufficient data on the scattering characteristics of marine targets, the Sea-Detecting Radar Data-Sharing Program (SDRDSP) was upgraded to obtain data on marine targets and their environment under different polarizations and sea states. This upgrade expanded the physical dimension of radar target observation and improved radar and auxiliary data acquisition capabilities. Furthermore, a dual-polarized multistate scattering characteristic dataset of marine targets was constructed, and the statistical distribution characteristics, time and space correlation, and Doppler spectrum were analyzed, supporting the data usage. In the future, the types and quantities of maritime targets will continue to accumulate, providing data support for improving marine target detection and recognition performance and intelligence.
Marine target detection and recognition depend on the characteristics of marine targets and sea clutter. Therefore, understanding the essential features of marine targets based on the measured data is crucial for advancing target detection and recognition technology. To address the issue of insufficient data on the scattering characteristics of marine targets, the Sea-Detecting Radar Data-Sharing Program (SDRDSP) was upgraded to obtain data on marine targets and their environment under different polarizations and sea states. This upgrade expanded the physical dimension of radar target observation and improved radar and auxiliary data acquisition capabilities. Furthermore, a dual-polarized multistate scattering characteristic dataset of marine targets was constructed, and the statistical distribution characteristics, time and space correlation, and Doppler spectrum were analyzed, supporting the data usage. In the future, the types and quantities of maritime targets will continue to accumulate, providing data support for improving marine target detection and recognition performance and intelligence.
3
2026, 15(2): 409-440.
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.
4
2026, 15(3): 996-1012.
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 image features 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 image features 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.
5
2025, 14(5): 1276-1293.
This study addresses the issue of fine-grained feature extraction and classification for Low-Slow-Small (LSS) targets, such as birds and drones, by proposing a multi-band multi-angle feature fusion classification method. First, data from five types of rotorcraft drones and bird models were collected at multiple angles using K-band and L-band frequency-modulated continuous-wave radars, forming a dataset for LSS target detection. Second, to capture the periodic vibration characteristics of the L-band target signals, empirical mode decomposition was applied to extract high-frequency features and reduce noise interference. For the K-band echo signals, short-time Fourier transform was applied to obtain high-resolution micro-Doppler features from various angles. Based on these features, a Multi-band Multi-angle Feature Fusion Network (MMFFNet) was designed, incorporating an improved convolutional long short-term memory network for temporal feature extraction, along with an attention fusion module and a multiscale feature fusion module. The proposed architecture improves target classification accuracy by integrating features from both bands and angles. Validation using a real-world dataset showed that compared with methods relying on single radar features, the proposed approach improved the classification accuracy for seven types of LSS targets by 3.1% under a high Signal-to-Noise Ratio (SNR) of 5 dB and by 12.3% under a low SNR of −3 dB.
This study addresses the issue of fine-grained feature extraction and classification for Low-Slow-Small (LSS) targets, such as birds and drones, by proposing a multi-band multi-angle feature fusion classification method. First, data from five types of rotorcraft drones and bird models were collected at multiple angles using K-band and L-band frequency-modulated continuous-wave radars, forming a dataset for LSS target detection. Second, to capture the periodic vibration characteristics of the L-band target signals, empirical mode decomposition was applied to extract high-frequency features and reduce noise interference. For the K-band echo signals, short-time Fourier transform was applied to obtain high-resolution micro-Doppler features from various angles. Based on these features, a Multi-band Multi-angle Feature Fusion Network (MMFFNet) was designed, incorporating an improved convolutional long short-term memory network for temporal feature extraction, along with an attention fusion module and a multiscale feature fusion module. The proposed architecture improves target classification accuracy by integrating features from both bands and angles. Validation using a real-world dataset showed that compared with methods relying on single radar features, the proposed approach improved the classification accuracy for seven types of LSS targets by 3.1% under a high Signal-to-Noise Ratio (SNR) of 5 dB and by 12.3% under a low SNR of −3 dB.
6
2026, 15(3): 983-995.
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.
7
2026, 15(3): 1059-1090.
Driven by complex electromagnetic environments and multi-target collaborative detection needs, enhancing the overall effectiveness of radar networks through autonomous coordination technology has become a key research area in radar collaborative surveillance. Extensive research has been conducted worldwide, yielding substantial advances in theoretical development, technical validation, and equipment application. This paper systematically discusses the foundational concepts and main features of autonomous coordination in radar networks, examining the primary technical challenges faced during implementation and performance optimization. It also reviews recent notable research findings and technological strategies, focusing on collaborative architecture design, sensing, intelligent decision-making and control, and autonomous evolution. Finally, this paper offers an outlook on future trends in the field and provides references for related theoretical research and practical applications.
Driven by complex electromagnetic environments and multi-target collaborative detection needs, enhancing the overall effectiveness of radar networks through autonomous coordination technology has become a key research area in radar collaborative surveillance. Extensive research has been conducted worldwide, yielding substantial advances in theoretical development, technical validation, and equipment application. This paper systematically discusses the foundational concepts and main features of autonomous coordination in radar networks, examining the primary technical challenges faced during implementation and performance optimization. It also reviews recent notable research findings and technological strategies, focusing on collaborative architecture design, sensing, intelligent decision-making and control, and autonomous evolution. Finally, this paper offers an outlook on future trends in the field and provides references for related theoretical research and practical applications.
8
,
The threat from low-altitude targets to airspace security such as airport is increasing, making accurate detection and recognition essential for radar systems. High-quality measured radar datasets are crucial for advancing low-altitude target recognition. However, most existing public radar datasets for these targets consist of simulation data or short-range collected data, which have difficulty accurately reflecting and verifying radar target recognition performance in long-range scenarios. To overcome these limitations, this study creates a low-altitude target detection and recognition dataset based on Holographic Staring Radar (HSR), including measured data collection and recognition validation for typical low-altitude targets in outdoor environments. The dataset includes common targets such as multirotor unmanned aerial vehicles, sparrows, and large migratory birds, along with representative motion scenarios like hovering, circling, and radial flight. It also offers synchronized target micro-Doppler waterfall plots and radar-measured track information (including azimuth and elevation angles, radial velocity, and normalized signal-to-noise ratio), providing a data foundation for exploring the intrinsic link between target detailed features and motion states. Building on this, a multimodal adaptive feature fusion network is developed to extract and combine Doppler and kinematic features from different targets, demonstrating the dataset’s effectiveness in distinguishing various low-altitude targets.
The threat from low-altitude targets to airspace security such as airport is increasing, making accurate detection and recognition essential for radar systems. High-quality measured radar datasets are crucial for advancing low-altitude target recognition. However, most existing public radar datasets for these targets consist of simulation data or short-range collected data, which have difficulty accurately reflecting and verifying radar target recognition performance in long-range scenarios. To overcome these limitations, this study creates a low-altitude target detection and recognition dataset based on Holographic Staring Radar (HSR), including measured data collection and recognition validation for typical low-altitude targets in outdoor environments. The dataset includes common targets such as multirotor unmanned aerial vehicles, sparrows, and large migratory birds, along with representative motion scenarios like hovering, circling, and radial flight. It also offers synchronized target micro-Doppler waterfall plots and radar-measured track information (including azimuth and elevation angles, radial velocity, and normalized signal-to-noise ratio), providing a data foundation for exploring the intrinsic link between target detailed features and motion states. Building on this, a multimodal adaptive feature fusion network is developed to extract and combine Doppler and kinematic features from different targets, demonstrating the dataset’s effectiveness in distinguishing various low-altitude targets.
9
To address issues such as insufficient feature extraction, limited spatiotemporal correlation modeling, and poor classification performance in radar classification of Low, Slow, and Small (LSS) 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 (LSS) 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.
10
2026, 15(3): 860-875.
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.
11
2023, 12(4): 906-922.
This study proposes a Synthetic Aperture Radar (SAR) aircraft detection and recognition method combined with scattering perception to address the problem of target discreteness and false alarms caused by strong background interference in SAR images. The global information is enhanced through a context-guided feature pyramid module, which suppresses strong disturbances in complex images and improves the accuracy of detection and recognition. Additionally, scatter key points are used to locate targets, and a scatter-aware detection module is designed to realize the fine correction of the regression boxes to improve target localization accuracy. This study generates and presents a high-resolution SAR-AIRcraft-1.0 dataset to verify the effectiveness of the proposed method and promote the research on SAR aircraft detection and recognition. The images in this dataset are obtained from the satellite Gaofen-3, which contains 4,368 images and 16,463 aircraft instances, covering seven aircraft categories, namely A220, A320/321, A330, ARJ21, Boeing737, Boeing787, and other. We apply the proposed method and common deep learning algorithms to the constructed dataset. The experimental results demonstrate the excellent effectiveness of our method combined with scattering perception. Furthermore, we establish benchmarks for the performance indicators of the dataset in different tasks such as SAR aircraft detection, recognition, and integrated detection and recognition.
This study proposes a Synthetic Aperture Radar (SAR) aircraft detection and recognition method combined with scattering perception to address the problem of target discreteness and false alarms caused by strong background interference in SAR images. The global information is enhanced through a context-guided feature pyramid module, which suppresses strong disturbances in complex images and improves the accuracy of detection and recognition. Additionally, scatter key points are used to locate targets, and a scatter-aware detection module is designed to realize the fine correction of the regression boxes to improve target localization accuracy. This study generates and presents a high-resolution SAR-AIRcraft-1.0 dataset to verify the effectiveness of the proposed method and promote the research on SAR aircraft detection and recognition. The images in this dataset are obtained from the satellite Gaofen-3, which contains 4,368 images and 16,463 aircraft instances, covering seven aircraft categories, namely A220, A320/321, A330, ARJ21, Boeing737, Boeing787, and other. We apply the proposed method and common deep learning algorithms to the constructed dataset. The experimental results demonstrate the excellent effectiveness of our method combined with scattering perception. Furthermore, we establish benchmarks for the performance indicators of the dataset in different tasks such as SAR aircraft detection, recognition, and integrated detection and recognition.
12
2026, 15(2): 583-604.
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.
13
2026, 15(3): 944-963.
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.
14
2025, 14(4): 1092-1114.
Compared to ground-based external radiation source radar, satellite signal-based external radiation source radar (i.e., satellite signal external radiation source radar) offers advantages such as global, all-time, and all-weather coverage, which can compensate for the limitations of ground-based external radiation source radar in terms of maritime coverage. In contrast to medium and high-altitude satellite signals, Low-Earth Orbit (LEO) communication satellite signals have advantages such as strong reception power and a large number of satellites, which can provide substantial detection range and accuracy for passive detection of maritime targets. In response to future development needs, this paper provides a detailed discussion of the research status and application prospects of satellite signal external radiation source radar, and presents a feasibility analysis for constructing a low-earth orbit communication satellite signal external radiation source radar system using Iridium and Starlink, two types of LEO communication satellite systems, which integrates high and low frequencies with both wide and narrow bandwidths. Based on this, the paper summarizes the technical challenges and potential solutions in the development of low-earth orbit communication satellite signal external radiation source radar systems. The aforementioned research can serve as an important reference for wide-area external radiation source radar detection.
Compared to ground-based external radiation source radar, satellite signal-based external radiation source radar (i.e., satellite signal external radiation source radar) offers advantages such as global, all-time, and all-weather coverage, which can compensate for the limitations of ground-based external radiation source radar in terms of maritime coverage. In contrast to medium and high-altitude satellite signals, Low-Earth Orbit (LEO) communication satellite signals have advantages such as strong reception power and a large number of satellites, which can provide substantial detection range and accuracy for passive detection of maritime targets. In response to future development needs, this paper provides a detailed discussion of the research status and application prospects of satellite signal external radiation source radar, and presents a feasibility analysis for constructing a low-earth orbit communication satellite signal external radiation source radar system using Iridium and Starlink, two types of LEO communication satellite systems, which integrates high and low frequencies with both wide and narrow bandwidths. Based on this, the paper summarizes the technical challenges and potential solutions in the development of low-earth orbit communication satellite signal external radiation source radar systems. The aforementioned research can serve as an important reference for wide-area external radiation source radar detection.
15
2026, 15(3): 876-890.
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.
16
2026, 15(3): 1027-1041.
Radar signal deinterleaving is a critical technology in electronic intelligence and electronic support measures systems. The classical histogram-based method, although valued for its simplicity, is susceptible to deceptive jamming under counter-reconnaissance parameter design. This study proposes a deinterleaving method that is resistant to such deception. The main contributions are as follows: a frame period detection mechanism compatible with pulse missing rates from 0% to 50% is established through theoretical derivation; by integrating autocorrelation and the overlap rate, accurate frame period identification is achieved, which effectively distinguishes interference disguised as fixed Pulse Repetition Intervals (PRI) and prevents interference with the deinterleaving process; moreover, a coherent discrimination mechanism is introduced to handle scenarios with similar parameters and to accommodate fixed, staggered, sliding, and wobulated PRI modulation—within a unified framework. Experimental results show that the performance of histogram-based methods degrades severely in the presence of counterreconnaissance parameters, with maximum performance dropping to 0, while the proposed method maintains a minimum performance of 96.5%. Meanwhile, the proposed method reaches a minimum performance of 95.31% in parameter-similar scenarios. The proposed method remains effective against the four modulation types, whether counterreconnaissance parameters are present or not. It demonstrates antideception capability against counterreconnaissance design, strong generalization across modulation types, and reliable performance in parameter-similar scenarios, thereby greatly improving the deinterleaving reliability in complex electromagnetic environments and offering important implications for the development of electronic warfare systems.
Radar signal deinterleaving is a critical technology in electronic intelligence and electronic support measures systems. The classical histogram-based method, although valued for its simplicity, is susceptible to deceptive jamming under counter-reconnaissance parameter design. This study proposes a deinterleaving method that is resistant to such deception. The main contributions are as follows: a frame period detection mechanism compatible with pulse missing rates from 0% to 50% is established through theoretical derivation; by integrating autocorrelation and the overlap rate, accurate frame period identification is achieved, which effectively distinguishes interference disguised as fixed Pulse Repetition Intervals (PRI) and prevents interference with the deinterleaving process; moreover, a coherent discrimination mechanism is introduced to handle scenarios with similar parameters and to accommodate fixed, staggered, sliding, and wobulated PRI modulation—within a unified framework. Experimental results show that the performance of histogram-based methods degrades severely in the presence of counterreconnaissance parameters, with maximum performance dropping to 0, while the proposed method maintains a minimum performance of 96.5%. Meanwhile, the proposed method reaches a minimum performance of 95.31% in parameter-similar scenarios. The proposed method remains effective against the four modulation types, whether counterreconnaissance parameters are present or not. It demonstrates antideception capability against counterreconnaissance design, strong generalization across modulation types, and reliable performance in parameter-similar scenarios, thereby greatly improving the deinterleaving reliability in complex electromagnetic environments and offering important implications for the development of electronic warfare systems.
17
2024, 13(3): 539-553.
Detection of small, slow-moving targets, such as drones using Unmanned Aerial Vehicles (UAVs) poses considerable challenges to radar target detection and recognition technology. There is an urgent need to establish relevant datasets to support the development and application of techniques for detecting small, slow-moving targets. This paper presents a dataset for detecting low-speed and small-size targets using a multiband Frequency Modulated Continuous Wave (FMCW) radar. The dataset utilizes Ku-band and L-band FMCW radar to collect echo data from six UAV types and exhibits diverse temporal and frequency domain resolutions and measurement capabilities by modulating radar cycles and bandwidth, generating an LSS-FMCWR-1.0 dataset (Low Slow Small, LSS). To further enhance the capability for extracting micro-Doppler features from UAVs, this paper proposes a method for UAV micro-Doppler extraction and parameter estimation based on the local maximum synchroextracting transform. Based on the Short Time Fourier Transform (STFT), this method extracts values at the maximum energy point in the time-frequency domain to retain useful signals and refine the time-frequency energy representation. Validation and analysis using the LSS-FMCWR-1.0 dataset demonstrate that this approach reduces entropy on an average by 5.3 dB and decreases estimation errors in rotor blade length by 27.7% compared with traditional time-frequency methods. Moreover, the proposed method provides the foundation for subsequent target recognition efforts because it balances high time-frequency resolution and parameter estimation capabilities.
Detection of small, slow-moving targets, such as drones using Unmanned Aerial Vehicles (UAVs) poses considerable challenges to radar target detection and recognition technology. There is an urgent need to establish relevant datasets to support the development and application of techniques for detecting small, slow-moving targets. This paper presents a dataset for detecting low-speed and small-size targets using a multiband Frequency Modulated Continuous Wave (FMCW) radar. The dataset utilizes Ku-band and L-band FMCW radar to collect echo data from six UAV types and exhibits diverse temporal and frequency domain resolutions and measurement capabilities by modulating radar cycles and bandwidth, generating an LSS-FMCWR-1.0 dataset (Low Slow Small, LSS). To further enhance the capability for extracting micro-Doppler features from UAVs, this paper proposes a method for UAV micro-Doppler extraction and parameter estimation based on the local maximum synchroextracting transform. Based on the Short Time Fourier Transform (STFT), this method extracts values at the maximum energy point in the time-frequency domain to retain useful signals and refine the time-frequency energy representation. Validation and analysis using the LSS-FMCWR-1.0 dataset demonstrate that this approach reduces entropy on an average by 5.3 dB and decreases estimation errors in rotor blade length by 27.7% compared with traditional time-frequency methods. Moreover, the proposed method provides the foundation for subsequent target recognition efforts because it balances high time-frequency resolution and parameter estimation capabilities.
18
2026, 15(3): 1013-1026.
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.
19
2026, 15(3): 964-982.
The spaceborne Hybrid-Polarimetric Synthetic Aperture Radar (HP-SAR) balances the acquisition of rich polarimetric information with high-performance imaging. It offers advantages such as low system complexity and reduced data acquisition costs, and has emerged as a prominent direction in multidimensional microwave remote sensing. LT-1 is China’s first radar satellite equipped with spaceborne HP imaging capability, and it is also the world’s first satellite to implement a multi-channel HP radar system. This study utilizes HP imagery from the LT-1 satellite to construct and systematically elaborate the HP-SAR Evaluation and Analytical Dataset (HEAD-1.0), thereby addressing the lack of high-quality open-source HP datasets. HEAD-1.0 aims to provide data support for the quantitative assessment of HP-SAR image quality, the development of HP-SAR technology, and the design of new satellite missions, with particular emphasis on supporting novel observational technologies for terrestrial, oceanic, and deep-space applications. It comprises three components: (1) LT-1 SAR imagery, including 30 HP-SAR images and 16 Quad-Polarimetric SAR (QP-SAR) images, covering an area of approximately 64000 km2; (2) Auxiliary data, including six optical images and Digital Elevation Model (DEM) data in the same area as SAR images; and (3) Annotation data, including 28 active/passive calibrators, approximately 17 km2 of land cover classification, and 23 polygonal/linear annotated planetary analog scenes. Based on HEAD-1.0, a preliminary qualitative and quantitative study was conducted, involving HP-SAR calibration, a comparison of terrain classification between HP-SAR and QP-SAR, and an analysis of HP characterizations of planetary analog scenes. In the future, an internationally advanced polarimetric SAR benchmark database will be constructed by integrating multi-platform, multi-band, multi-angle, and multi-temporal imaging data. In particular, the future study will focus on supporting innovative research on key technologies, including planetary surface and subsurface exploration, intelligent fusion of multisource remote sensing data, and advanced interpretation algorithms for SAR imagery.
The spaceborne Hybrid-Polarimetric Synthetic Aperture Radar (HP-SAR) balances the acquisition of rich polarimetric information with high-performance imaging. It offers advantages such as low system complexity and reduced data acquisition costs, and has emerged as a prominent direction in multidimensional microwave remote sensing. LT-1 is China’s first radar satellite equipped with spaceborne HP imaging capability, and it is also the world’s first satellite to implement a multi-channel HP radar system. This study utilizes HP imagery from the LT-1 satellite to construct and systematically elaborate the HP-SAR Evaluation and Analytical Dataset (HEAD-1.0), thereby addressing the lack of high-quality open-source HP datasets. HEAD-1.0 aims to provide data support for the quantitative assessment of HP-SAR image quality, the development of HP-SAR technology, and the design of new satellite missions, with particular emphasis on supporting novel observational technologies for terrestrial, oceanic, and deep-space applications. It comprises three components: (1) LT-1 SAR imagery, including 30 HP-SAR images and 16 Quad-Polarimetric SAR (QP-SAR) images, covering an area of approximately 64000 km2; (2) Auxiliary data, including six optical images and Digital Elevation Model (DEM) data in the same area as SAR images; and (3) Annotation data, including 28 active/passive calibrators, approximately 17 km2 of land cover classification, and 23 polygonal/linear annotated planetary analog scenes. Based on HEAD-1.0, a preliminary qualitative and quantitative study was conducted, involving HP-SAR calibration, a comparison of terrain classification between HP-SAR and QP-SAR, and an analysis of HP characterizations of planetary analog scenes. In the future, an internationally advanced polarimetric SAR benchmark database will be constructed by integrating multi-platform, multi-band, multi-angle, and multi-temporal imaging data. In particular, the future study will focus on supporting innovative research on key technologies, including planetary surface and subsurface exploration, intelligent fusion of multisource remote sensing data, and advanced interpretation algorithms for SAR imagery.
20
2026, 15(3): 1091-1105.
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.
- First
- Prev
- 1
- 2
- 3
- 4
- 5
- Next
- Last
- Total:5
- To
- Go
微信 | 公众平台 
