Identifying satellite components in polarimetric inverse synthetic aperture radar (ISAR) images is beneficial for monitoring their operation and health status. Most target recognition methods rely on network structures designed for optical images and fail to consider the inherent polarimetric scattering characteristics. Furthermore, the aliasing of scattering mechanisms caused by the complex structure of man-made targets, along with the scattering diversity resulting from observation perspectives, poses challenges to target polarimetric interpretation. To address these challenges, this study proposes a structure recognition framework embedded within scattering mechanism to achieve pixel-level to component-level structure (CS) recognition. First, through semi-supervised representation learning, the 3-D polarimetric correlation pattern (3-D PCP) of typical polarimetric scattering structures (PSSs) is used as expert knowledge to guide a deep-learning network, enabling pixel-level scattering mechanism separation. On this basis, a relation module is employed to explore the relationships between different pixels’ scattering mechanisms to accomplish component-level recognition. Finally, polarimetric ISAR satellite images and component annotation datasets are constructed. Pixel-level and component-level comparisons verify the advantages of the proposed method.

ISAR enables continuous space surveillance irrespective of weather conditions, while the compact polarization (CP) mode balances hardware costs with the provision of polarization information. Recognizing components of space targets provides valuable insights into attitude inversion and the detection of abnormal motion. However, existing ISAR space target recognition methods lack the capability to transition from target-level classification to component-level recognition. Additionally, components with weak and non-uniform scattering intensity distributions pose challenges to their precise positioning and classification in ISAR images. To address these limitations, a coarse-to-fine dual-branch correlation aggregation graph attention network (DCA-Net) is proposed, featuring a novel graph neural network construction strategy. Considering target sparsity, a multi-channel graph node filter (MGNF) module combining compact polarimetric features is devised to enhance computation efficiency. Subsequently, a dual-branch correlation aggregation graph (DCAG) module is constructed concerning the local correlation and global topology. Information suppressed by the non-maximum suppression (NMS) algorithm is reutilized to construct a local graph, which is then aggregated to improve the recognition probability of weak scattering intensity components. Meanwhile, a global graph is constructed, allowing the utilization of structural topology relationships for robust inference in heterogeneous scattering scenarios. Experimental results on ISAR dataset demonstrate that the proposed method achieves superior performance, with at least a 5.38% improvement in the F1 score index and 5.21% improvement in the mean average precision (mAP) index.

Recently, integrated deep learning methods cascading speckle filtering and target detection have garnered increasing attention in synthetic aperture radar (SAR) image target detection. These methods use despeckling networks to suppress speckle noise, enhancing the performance of followed detectors. However, recent studies have shown that deep neural networks (DNNs) are vulnerable to adversarial attacks, where adding imperceptible perturbations to benign examples can cause incorrect predictions. This phenomenon raises serious concerns about the security of current integrated deep learning algorithms in SAR images. To this end, this work conducts adversarial attack research on integrated speckle filtering and detection methods to assess its adversarial robustness. Specifically, a novel meta-adversarial despeckling network (Meta-ADNet) architecture is proposed, which leverages the despeckling network as a potential attack pathway to inject perturbations, generating adversarial despeckled examples that invalidate subsequent detection processes. Meta-ADNet consists of two core components: the baseline model adversarial despeckling network (ADNet) and the corresponding meta-adversarial attack framework. ADNet first generates benign despeckled SAR images through a despeckling network, which are then fed into the perturbation injection branch. This branch guides the despeckling network in producing adversarial examples by backpropagating the confidence loss of the surrogate detection model. The meta-adversarial attack framework constructs different tasks by selecting multiple surrogate detection models and iteratively simulates white-box ensemble attacks and black-box attacks within each task to enhance the transfer attack capability of ADNet on black-box models. Extensive experimental results on the SAR detection datasets SSDD and HRSID demonstrate that the proposed algorithm can effectively attack white-box surrogate detectors and exhibits strong black-box transfer attack capabilities.

Stepped frequency chirp signal obtains high-resolution radar images by synthesizing multiple narrowband chirp pulses. It has been one of the most commonly used wideband radar waveforms due to its lower demand for radar instant bandwidth. In this paper, we propose a radar jamming method using two-dimensional nonperiodic phase modulation against stepped frequency chirp signal imaging radar. Using the unique property of nonperiodic phase modulation, the proposed method can generate high-level sidelobes that perform as a special blanket jamming along both the range and azimuth directions and make the target unrecognizable. Then, the influence of different modulation parameters, such as the code width and duty ratio, are further discussed. Based on this, the corresponding parameter design principles are presented. Finally, the validity of the proposed method is demonstrated by the Yake-42 plane data simulation and measured unmanned aerial vehicle data experiment.

Forward scatter radar (FSR) can achieve target detection by capturing the signal disturbance caused by targets crossing the baseline, which provides a countermeasure to stealth/small targets. Firstly, the FSR net model based on the Global Navigation Satellite System (GNSS) is constructed. Then, the statistical features in the time and frequency domains are used to describe the fluctuation characteristics of the forward-scatter signal. Moreover, the influences of the relative position of the target and baseline on the statistical features are analyzed through dynamic simulations, and the change regulation of time and frequency features is revealed. Finally, a FSR experiment is carried out in an anechoic chamber and the results validate the theoretical analysis. This work provides a meaningful reference for detection and parameter estimation of targets crossing the baseline in a FSR.

Compared with the optical imaging system, polarimetric inverse synthetic aperture radar (ISAR) can work all-day and all-weather, which plays an important role in space surveillance. However, high-resolution (HR) ISAR images usually require large bandwidth and coherent integration angle, which is limited by the equipment’s physical conditions. In this vein, the super-resolution (SR) of ISAR images is of vital importance. At present, supervised learning methods are often used in image SR of computer vision. By constructing low-resolution (LR) and HR data pairs, the neural network can learn the mapping relationship between them. However, the low-frequency information in LR image data is less considered. In addition, to obtain different scales of SR reconstruction results, multiple network training repetitions are usually needed, which consumes time and hardware resources. Based on the idea of implicit neural representation, this paper constructs an implicit neural network representation framework for polarimetric ISAR image SR, which can obtain multiscale SR results through one training. A semi-supervised module is also constructed to make the network have the ability of supervised and unsupervised learning, which is conducive to mine and make better use of LR images. A polarimetric ISAR image SR dataset is constructed for satellite targets while four indexes are adopted for quantitative evaluation in global and local aspects. Experiments demonstrate that the proposed approach achieves better SR performance, where the PSNR index can be increased at least by 0.93dB.

Polarimetric inverse synthetic aperture radar (ISAR) can operate all-day and all-weather, making it crucial for space surveillance. The compact polarimetric mode balances hardware complexity and polarimetric information, which is commonly equipped with ISAR systems. Given the constraints of limited physical conditions, exploring ISAR image super-resolution is worthwhile. Currently, deep learning models have been employed for enhancing ISAR image super-resolution. However, the super-resolution performance is limited by local interpolation and the occurrence of artifacts. To address these limitations, this work presents a Non-Local Scene Awareness Network (NLSAN), which incorporates a non-local interpolation approach to capture global textures. Furthermore, a scene awareness scheme is established by integrating semantic and super-resolution information, concerning the varying levels of artifacts in different regions. The training process can be regulated by a designed penalty function to mitigate potentially generated artifacts. A dataset of compact polarimetric ISAR images of satellite targets is constructed for comparison analysis. The proposed NLSAN method yields more elaborate super-resolution results with fewer artifacts. Quantitative evaluations are also carried out using global and local indexes such as the Peak-Signal-to-Noise (PSNR), the image entropy, and the 3dB width of strong scatters. Compared with the typical state-of-the-art methods, the proposed approach achieves superior super-resolution performance, with an overall performance improvement of at least 9.2% and enhanced generalization capabilities.

This letter proposes a novel algorithm to estimate the 3-D attitude using bistatic full-polarization information. First, tilt angles of bistatic Huynen target parameters are extracted from the Sinclair matrix. Then, extracted tilt angles are used to recover the linear-type target attitude by parametric space searching and noncoherent accumulation. Anechoic chamber experiment results demonstrate that the max relative error is smaller than 3% for an iron wire object with two attitude placements.

Polarimetric radar can acquire complete polarization information and is widely used in many applications. However, target orientation relative to the radar line of sight usually exhibits significant influences on the scattering mechanisms. Recently, such target scattering diversity has been successfully characterized and utilized with the polarimetric rotation domain interpretation techniques. In radar polarimetry, target scattering responses are affected by both polarization orientation angle and polarization ellipticity angle. In this vein, this work aims at exploring and utilizing the complete target scattering diversity by extending polarimetric rotation domain techniques to the polarization ellipticity angle dimension. The main idea is to develop a three-dimension polarimetric correlation pattern (3-D PCP) interpretation tool, which can visualize and exhibit targets’ polarimetric rotation domain properties in terms of both the polarimetric orientation and ellipticity angles. Then, a set of global, local, and mutual polarimetric features are proposed to characterize the responses of a 3-D PCP interpretation tool. Especially, the curvatures in differential geometry are first introduced to describe the properties of the 3-D surface. The performance of these new polarimetric features is investigated with spaceborne polarimetric synthetic aperture radar (PolSAR) data. Experimental results demonstrate the advantage of the proposed 3-D PCP features in enhancing the target clutter ratio (TCR), especially for the weak ship area. Based on this, a non-local superpixel-level contrast measure (NSLCM) method for ship detection is proposed. Pure sea samples can be determined adaptively for salient map construction. Comparison results demonstrate better detection performance for both the inshore dense ship area and weak ship area.

Model-based decompositions are powerful tools for scattering mechanism interpretation of polarimetric synthetic aperture radar (PolSAR) data. By incorporating their refined physical scattering models and utilizing the excellent nonlinear data fitting capability of neural networks, a target-to-mechanism mapping network is proposed. Inputting the polarimetric features defined in the normalized polarimetric feature space, the proposed network outputs the normalized powers of four scattering components. Experimental studies demonstrate that the trained network on Pi-SAR X-band PolSAR data shows good interpretation performance and generality on the cross-observation perspective Pi-SAR X-band PolSAR data and the cross-frequency Radardat-2 C-band PolSAR data. In addition, the proposed approach has a fast interpretation speed.

Speckle filtering of polarimetric synthetic aperture radar (PolSAR) data is a necessary preprocessing step for many subsequent applications. The performances of speckle reduction and details preservation are primarily determined by the scheme of similar pixel selection. This letter mainly contributes to establish a novel approach for adaptive and efficient selection of similar pixels. The core idea is to introduce the clustering concept to collect similar samples during the similarity test of polarimetric matrices which is functioned to be a distance measure. Adaptive clustering based on fast finding of density peaks of the data elements is adapted and an automatic determination strategy of cluster numbers is developed. Then the proposed speckle filter is established based on the distance measure and the adaptive clustering. Both unmanned aerial vehicle SAR (UAVSAR) and Radarsat-2 data sets are used for experimental studies. The demonstrations of similar pixel selection for several typical scattering patterns are carried out. The similar samples have been adaptively and precisely identified by the proposed scheme. Furthermore, the comparisons of speckle filtering performances with several advanced speckle filters clearly demonstrate the efficiency and superiority of the proposed method.

Due to the high cost of obtaining echo data for space targets in real cases, most of the research usually relies on the simulation. The electromagnetic (EM) calculation is well known as one of the simulation tools. In this letter, the relationship between the calculated complex radar cross-section data and the baseband echo signal is analyzed. Then, the difference in the polarization reference plane of the radar coordinate system and EM calculation coordinate system is pointed out. For simulating the baseband echo signal with full polarization, the scattering matrix obtained by EM calculation should be rotated according to the radar position, the target position, and the target attitude. At last, a simulation of the baseband echo signal with full polarization for a cone-shaped target is given to verify the simulation algorithm.

Ship detection is one of important applications for synthetic aperture radar (SAR). Speckle effects usually make SAR image understanding difficult and speckle reduction becomes a necessary pre-processing step for majority SAR applications. This work examines different speckle reduction methods on SAR ship detection performances. It is found out that the influences of different speckle filters are significant which can be positive or negative. However, how to select a suitable combination of speckle filters and ship detectors is lack of theoretical basis and is also data-orientated. To overcome this limitation, a speckle-free SAR ship detection approach is proposed. A similar pixel number (SPN) indicator which can effectively identify salient target is derived, during the similar pixel selection procedure with the context covariance matrix (CCM) similarity test. The underlying principle lies in that ship and sea clutter candidates show different properties of homogeneity within a moving window and the SPN indicator can clearly reflect their differences. The sensitivity and efficiency of the SPN indicator is examined and demonstrated. Then, a speckle-free SAR ship detection approach is established based on the SPN indicator. The detection flowchart is also given. Experimental and comparison studies are carried out with three kinds of spaceborne SAR datasets in terms of different polarizations. The proposed method achieves the best SAR ship detection performances with the highest figures of merits (FoM) of 97.14%, 90.32% and 93.75% for the used Radarsat-2, GaoFen-3 and Sentinel-1 datasets, accordingly.

Speckle reduction and target detection are usually two independent and successive procedures in SAR information processing systems. The separated implementation scheme makes the determination of a suitable combination of speckle filter and target detector a challenging task in practice. This work attempts to propose a novel integrated approach for both SAR speckle reduction and target detection. Firstly, a new representation in terms of the context scattering vector and context covariance matrix is established for information augmentation and mining for SAR data. Then, similar pixels within a large moving window are selected using matrix similarity test and the similar pixel number (SPN) is recorded. The speckle reduction can be conducted with the determined similar samples and salient targets (e.g. manmade targets) can be detected with the SPN index simultaneously. Finally, the speckle reduction and ship detection integrated processing procedure is established. Experimental studies are carried out with space-borne SAR datasets. The proposed integrated framework shows superiority in both speckle reduction and target detection performances.

Target classification is a main application of polarimetric synthetic aperture radar (PolSAR). With the PolSAR images of the same region observed on various dates, multitemporal classification plays an important role in crop discrimination and growth monitoring. However, due to the crop growth change, classification methods with good generalization are imperative. Recently, advanced deep learning techniques have achieved breakthroughs in optical image processing. Among them, convolutional neural networks (CNNs) can greatly improve accuracies in classification tasks. Because their multilayer architectures can extract the abstract feature with strong robustness to characterize the kinds' difference. In this vein, the multilayer architectures may be able to lead to the higher accuracies, as well as the better generalization in multitemporal PolSAR classification. So, this work aims to investigate multitemporal PolSAR classification using five CNNs of AlexNet, VGG16, Inception, ResNet50, and MobileNet. In detail, four roll-invariant features and two hidden features in the rotation domain are selected as inputs. Comparison experiment based on four temporal UAVSAR data with seven land covers validates the efficiency of these CNNs. For the train-used temporal, Inception can achieve the higher overall accuracies. While, for the train-not-used temporal, ResNet50 is with the better generalization.

The low-cost sensor array, an effective way to provide sufficient degrees of freedom (DoFs) and affordable costs, is one of the most important foundation devices for large-scale applications of the Internet of Things (IoT) and sixth-generation (6G) spaceborne communication systems. In this article, we propose an irregular subarray-based low-cost sensor array design strategy, which decomposes the practical application problem into two subproblems according to the scene scale. For the small-size cases, a user-defined DoF control strategy is presented, which is flexibly defined and establishes an effective tradeoff between the sensor array radiation performance and manufacturing cost. Moreover, for large-size cases, a hierarchical subarray design strategy is presented. A module panel structure composed of subarray tiles is proposed to avoid both the convergence problem in high-dimensional optimization and engineering complexity. Both design strategies are able to ensure good sensor radiation performance while maintaining engineering convenience. The effectiveness and potential of the proposed method are verified by various numerical examples.

Roll-invariant polarimetric features which are independent of target orientation angle along the radar line of sight are popularly adopted in many radar application fields. During the development history of radar polarimetry, a number of roll-invariant polarimetric features have been reported in terms of different polarimetric matrix formulations, different target decomposition approaches and different scattering mechanism interpretation tools. Currently, there is lack of a comprehensive summary of these valuable roll-invariant polarimetric features. Also, their inner-relationships need to be further disclosed. Finally, deep investigations of their application potentials are essentially necessary. This work is dedicated to these aforementioned issues and a survey of roll-invariant polarimetric features is carried out.

Sea clutter submerges the small target echo, which is disadvantageous for radar target detection. In this letter, we propose to suppress the sea clutter by orthogonal projection (OP). We construct the clutter subspace directly by the observed data vectors of neighboring range cells of the cell under test (CUT), and then suppress the sea clutter of the CUT by projecting the observed signal of the CUT to the clutter orthogonal subspace. We design a new detector which combines the cell averaging constant false alarm rate (CFAR) detector with the OP. The clutter suppression performance and computation complexity of OP are compared with those of singular value decomposition (SVD), respectively. Theoretical analysis and experimental results with real sea clutter demonstrate that the radar CFAR detection performance improves by using OP. And the CFAR detector based on OP is convenient to implement. The CFAR detection performance based on OP is the same as that based on SVD, whereas the computation complexity of OP is much less than that of SVD.

Deep convolutional neural network (CNN) techniques have been utilized to enhance polarimetric synthetic aperture radar (PolSAR) image classification performance. This work contributes to a current challenge that is how to adapt deep CNN classifier for PolSAR classification with limited training samples while keeping good generalization performance. A polarimetric-feature-driven deep CNN classification scheme is established with both classical roll-invariant polarimetric features and hidden polarimetric features in the rotation domain to drive the proposed deep CNN model. Comparison studies validate the efficiency and superiority of the proposal. For the benchmark AIRSAR data, the proposed method achieves the state-of-the-art classification accuracies. Meanwhile, the convergence speed from the proposed CNN approach is about 2.3 times faster than the normal CNN method. For multi-temporal UAVSAR datasets, the proposed scheme achieves comparably high classification accuracies as the normal CNN method for train-used temporal data, while for train-not-used data it obtains average 4.86% higher overall accuracy than the normal CNN method. Furthermore, the proposed strategy can also produce very promising classification accuracy with very limited training samples.

Feature extraction and recognition of wobbling targets are very important in spatial target surveillance. It has been shown that the scattering centers are slipping on the edge of the rotational symmetry target with the target micro-motion according to the electromagnetic scattering theory and electromagnetic computation analysis. Based on the slippery scattering center model, the micro-motion model and high-range resolution profile (HRRP) model of a wobbling rotational symmetry cone-shaped target are introduced, and the observed HRRP sequence is used to construct a time-range distribution matrix, then a estimation method of the wobbling period is proposed based on time-range distribution matrix correlation, which is validated by electromagnetic computation data and dynamic simulation.

Orientation angle compensation was incorporated into model-based decomposition to cure overestimation of the volume scattering contribution for interpretation of polarimetric synthetic aperture radar (PolSAR) data. The compensation is based on rotating the coherency matrix to minimize the cross-polarization term. However, this processing cannot always guarantee that the double- and odd-bounce scattering components will be rotated back to zero orientation angle and left with zero cross-polarization power. As a result, built-up patches with large orientation angles may still suffer from the scattering mechanism ambiguity. In this paper, double- and odd-bounce scattering models were generalized to fit the cross-polarization and off-diagonal terms, by separating their independent orientation angles. A general decomposition framework is proposed that utilizes all elements of a coherency matrix. The residual minimization criterion is used for model inversion. All the model parameters are simultaneously obtained using a nonlinear least squares optimization technique. The manual intervention, branch conditions, and negative power issues are avoided. The performance and advantages of this approach are demonstrated and evaluated with spaceborne L-band ALOS/PALSAR and airborne X-band Pi-SAR PolSAR data sets. Comparison studies are also carried out and demonstrate that further improved decomposition performance is achieved by the proposed method, especially in oriented built-up areas.

This paper presents the development of a uniform polarimetric matrix rotation theory in the rotation domain along the radar line of sight for polarimetric SAR (PolSAR) data interpretation. The uniform representation of each coherency matrix element is a sinusoidal function in the rotation domain. A set of oscillation parameters, including oscillation amplitude, oscillation center, angular frequency and initial angle, is proposed to fully characterize the scattering behavior in the rotation domain. A set of rotation angle parameters, including stationary angle, null angle, and minimization/maximization angles, is derived from the angular frequency and initial angle to indicate the specific states of the rotation property. A look-up table for these parameters is provided and their physical meanings are interpreted. The proposed theory generalizes both the classic polarization orientation (PO) angle originally derived from the covariance matrix in a circular polarization basis and the deorientation theory developed from the minimization of the cross-polarization term. The roll-invariant terms have also been summarized. Finally, multi-frequency AIRSAR and Pi-SAR PolSAR data sets are used to demonstrate the derived parameters.

The 2-D image of a spinning target obtained by monostatic radar is modified by the angle between the spinning axis and the radar line of sight, so it usually cannot describe the real size of the target. The T/R-R bistatic radar is proposed to solve this problem. First, the bistatic geometry and bistatic echo signals of a spinning target are modeled. Then, a novel 2-D imaging algorithm is presented based on the connections between the mono- and bistatic echoes of the same scatterer and the classical Hough transform, allowing both the mono- and bistatic 2-D images of the spinning target to be obtained simultaneously. The theoretical analysis is carried out by the electromagnetic calculation and numerical simulations.

A new cross-channel interference suppression method based on the fractional Fourier transform (FRFT) is proposed to eliminate the effect due to cross-channel interference when a full-polarimetric radar adopts the opposite-slope linear frequency modulation transmit signals and de-ramping processing in the simultaneous measurement mode. The FRFT with an appropriate order is operated on the signals in the cross-interference interval. Such order is obtained via the calculation of the frequency-modulation slope of transmitted signals. The cross-channel interference is changed into sinc functions that are detected by a designed threshold and filtered out in the FRFT domain to suppress the cross-channel interference. The simulations show that the proposed method can obviously reduce the peak sidelobe levels and the integrated sidelobe levels of targets in range profiles compared with the traditional method, which is benefit to improve the detection and measurement performances of weak polarimetric scattering components. The proposed method is suitable for stable or slow-moving targets.

An extraction method of the polarimetric atmospheric target has been proposed to obtain most useful information with noise, clutter and artificial signals as less as possible for dual- orthogonal frequency modulation continuous wave (FMCW) polarimetric radar. The signal processing is described for dual- orthogonal FMCW polarimetric radar. The polarimetric range- Doppler spectra of precipitation obtained in a single sweep time are calibrated by radar constant calibration and reciprocity modification. The mask matrix for filtering is constructed based on the double spectral linear depolarization ratios (sLdr) and mathematical morphology methods. The precipitation atmospheric targets are extracted successfully by multiplying the mask matrix with the measured range-Doppler spectra by the polarimetric agile radar SAnd X-band (PARSAX) radar. The proposed technique keeps more atmospheric targets and suppresses more clutter, noise and artificial signals compared with the original data and the data after double sLdr filtering combined with noise clipping.