With multiple channels receiving echoes in azimuth, multichannel SAR has the potential of forward-looking imaging. However, its angular resolution, especially the area along the platform flight path, is poor due to the small variation of viewing angle. In this paper, the scheme of angular resolution enhancement for multichannel forward-looking SAR imaging based on zero-shot learning is proposed. In the scheme, the preliminary imaging results are obtained firstly with the synthetic aperture processing, and then the results are divided into three equal parts in azimuth, namely left, middle and right parts, to construct the dataset. Then by incorporating the frequency domain loss and mixed attention module, the modified CycleGAN network is developed to learning the mapping between the middle part (the area with lower resolution) and the left/right parts (the area with higher resolution). At last, experimental results are illustrated to verify that the proposed scheme can enhance the angular resolution of the middle area to the level of left/right area for the forward-looking images even without ground truths.

Due to the high degree of freedom of bistatic synthetic aperture radar (SAR), how to design a suitable configuration to achieve forward-looking high-quality imaging is a crucial issue. Since the spatial resolution is the most critical performance metric, the configuration design of bistatic forward-looking SAR (BFSAR) driven by spatial resolution metrics is discussed in this letter. First, based on the general geometry configuration and generalized ambiguity function (GAF), the analytical expression of the ellipse resolution for BFSAR is derived. Then, a novel measure for spatial resolution is developed by considering both the orthogonality and balance. Finally, with the prior information of the transmitting platform, the optimal configuration of the forward-looking receiver is established by solving an optimization problem of the spatial resolution with genetic algorithm, and simulation results are illustrated to validate the effectiveness of the proposed method.

The reconstruction of the three-dimensional structure of synthetic aperture radar (SAR) targets is a hot and difficult issue in the field of SAR. Conventional 3-D reconstruction methods based on 2-D SAR images do not consider the inherent characteristics of SAR imaging such as geometric deformation, overlap, and occlusion, and can only reconstruct simple and regular targets. To address this, we propose a CNN-based SAR 3-D reconstruction method called Multi-aspect SAR Stereo Network (MASS-Net). Our network is an end-to-end deep learning architecture that can automatically complete dense matching among multi-aspect SAR images and calculate the height to obtain height maps by learning prior knowledge. In the network, a feature extractor based on CNN is constructed to extract features from SAR images, which can extract high-dimensional features of 2-D SAR images, and help to capture neighborhood information and overcome the influence of geometric deformation and occlusion. Then a differentiable SAR projection relationship is established to construct a cost volume that includes the differences in multi-aspect image features. This projection relationship ensures ensures the overall differentiability of our pipeline. Meanwhile, the encoding and decoding architecture based on 3-D CNN is utilized to achieve regularization and regression to generate height maps. Finally, we use multi-aspect height maps to construct a dense 3D point cloud of the target. These make MASS-Net efficient and effective. Compared with traditional methods, our method can address issues such as distortion and occlusion, and efficiently reconstruct dense and accurate 3-D point cloud of complex targets. Simulation experiments and actual measurement experiments have been conducted to verify our proposed method.

Multichannel synthetic aperture radar (SAR) has the potential of resolving left/right ambiguity and then achieving high-resolution forward-looking imaging. However, when the sparse channel configuration is adopted, there are always residual grating lobes and left/right ambiguity in the imaging results. In this paper, a scheme of simultaneously suppressing residual grating lobes and left/right ambiguity for sparse channel forward-looking SAR imaging is proposed. Firstly, the formation of the residual grating lobes and left/right ambiguity are analyzed based on the signal model and imaging procedure of multichannel forward-looking SAR. Then, the characteristics of the differences in the position of grating lobes for the imaging results of different snapshots, and the spatiotemporal coupling characteristics of suppression of grating lobes and left/right ambiguity are explained. Next, a space-time steering matrix based on range history information is constructed, which establishes a linear mapping relationship between the space-time echo and the original scene. By solving the linear equation with a regularized iterative adaptive approach, the residual grating lobes and left/right ambiguity are suppressed simultaneously. The extensive simulation and experimental results demonstrate the effectiveness of the proposed method.

With the increasing number of electromagnetic devices, radio frequency interference (RFI) suppression has gradually become an essential problem in synthetic aperture radar (SAR) imaging. Faced with complex RFI environments such as time-varying and multitype mixing that may occur, traditional approaches often result in inadequate suppression and loss of valuable echoes. Moreover, the representation ability of manually extracted features is limited, struggling to maintain consistent performance in complex electromagnetic environments. To tackle these challenges, this article proposes a feature-enhanced low-rank and sparse decomposition network (FELS-Net), which separates RFI and useful echoes in the time-frequency domain (TFD). We introduce two learnable invertible nonlinear transforms to enhance the representation of RFI and SAR echoes, and unfold the RFI suppression scheme based on low-rank and sparse decomposition into a parameter-learnable network structure. The strong interpretability of model-driven architecture offers a potential stability guarantee for RFI suppression performance, while deep learning (DL) contributes to more effective feature characterization and more efficient and robust parameter schemes. Experimental results demonstrate that the proposed method outperforms comparative approaches in both time-frequency (TF) and image domains, exhibiting robust performance across diverse experimental conditions.

Real aperture radar superresolution imaging of sea surface targets has significant applications in sea surface surveillance and maritime rescue. However, its performance rapidly deteriorates when the echo data of the valid targets are merged into the strong sea clutter. To address this challenge, this paper proposes a multi-feature divide-and-conquer (MF-D&C) framework by forming a complex multi-feature enhancement network (CMFE-NET) and data-divide-and-conquer-based (DD&C-based) sparse Bayesian learning (SBL) algorithm. First, to separate sea clutter echo from valid targets’ echo, a CMFE-NET is proposed to transform the complex echoes into four distinct feature spaces: amplitude, phase, frequency, and dwell time. Second, based on the separated sea clutter echo and the valid targets’ echo, a DD&C-based SBL algorithm is proposed to perform Bayesian parameter estimation on both the clutter and target components, which improves the model parameter estimation performance within the Bayesian framework. Finally, a parameter pruning solver is introduced in EM estimation to eliminate inactive parameters during the iterative super-resolution process, significantly reducing computational overhead. The proposed framework demonstrates superior capabilities resolution enhancement in sea surface target superresolution imaging.

The reconstruction of the scale and shape information of extended targets is a major challenge for real aperture radar. Due to the limitation of angular resolution, the reconstruction of extended targets is inaccurate. To this end, a sparse and two-dimensional high-order gradient (S-2DHG) hybrid prior-based Bayesian method was proposed for real aperture radar to reconstruct the extended targets by introducing a novel scale-constrained prior into the framework of existing hybrid priors. On the one hand, the proposed 2DHG prior establishes interconnections among multiple adjacent units in both the range and azimuth directions during the reconstruction of the scattering coefficient unit and the current target. This interconnection facilitates the formation of a 2DHG prior, which effectively mitigates the influence of sidelobes in both range and azimuth. The sparse prior helps to alleviate the resolution loss of the 2DHG prior. On the other hand, the proposed Bayesian solution framework introduces Jeffery uninformative prior, which can realize the adaptive update of sparse scale prior weight parameters, reducing the number of manually selected parameters. Simulation and experimental results present superior data fidelity and edge preservation ability of the proposed method, which can accurately reconstruct the scale information of the extended targets.

Bistatic Synthetic Aperture Radar (BiSAR) enables highly flexible configuration, offering broad application prospects. However, existing BiSAR imaging algorithms neglect the complex scattering characteristics of the target, resulting in the loss of target structural information in the imaging results. To enhance the target structural information in imaging results and improve the interpretability of BiSAR images, we first establish the BiSAR echo model based on the target scattering model and analyze the target’s imaging characteristics by incorporating the imaging mechanisms. Subsequently, based on the imaging characteristics, we propose a structure-driven multi-stage BiSAR trajectory planning method (SMTP). This method solves a multi-stage multi-objective optimization problem driven by target scattering characteristics, thereby fully presenting all discernible structural features in the imaging results. Numerical simulation experiments validate the proposed method, demonstrating its ability to recover target structural information. This approach addresses the gap where BiSAR mission planning has largely overlooked target-specific characteristics.

Range sidelobe is a common and widely concerned issue in synthetic aperture radar (SAR) imaging. The sidelobe of strong scatters could cover weak targets and reduce the quality of SAR images, increasing the difficulty in interpretation. Due to the principle of energy conservation, reducing the sidelobe will reduce the range resolution of SAR for the single waveform design. By taking full advantage of the degrees of freedom of the transmitter, transmitting multiple waveforms, and exploiting the information discrepancy between pulses, we can overcome this limitation. Complementary sequences are a typical example since the range sidelobe of the sequences can be theoretically summed up to zero. Due to the transmission of different waveforms between pulses, an amplitude-phase modulation will be introduced in the azimuth dimension of SAR echo, which, in turn, leads to the Doppler spectrum aliasing and unexpected energy spikes outside the main energy region of SAR point spread function (PSF). Seeking to suppress the range sidelobe and meanwhile mitigate the spikes caused by waveform agility in SAR, we first establish the expression of the echo under the agile transmitting mode. Then, we propose an SAR PSF shaping (SAR-PSFS) method to suppress the integrated sidelobe level (ISL) of the whole range sidelobe plane in the SAR image via complementary waveform optimization. The inexact alternating direction penalty method (IADPM) framework is adopted to solve the resulting nonconvex optimization problem. Simulation results show that the proposed method outperforms the nonlinear frequency-modulation (NLFM) signal with a 5.46-dB lower integrated sidelobe ratio (ISLR) under the same mainlobe width. Besides, the range sidelobe can be effectively canceled with a 3.26-dB lower peak sidelobe ratio (PSLR) and a 4.99-dB lower ISLR compared with the Hanning window while maintaining the range resolution with the spikes caused by waveform agility effectively mitigated.

Multistatic three-dimensional synthetic aperture radar (3D SAR) has shown significant potential in rapid 3D imaging. Compared to traditional multi-pass or array 3D imaging systems, it achieves high-resolution imaging in a single pass. However, due to the introduction of multiple radar systems, decoherence factors such as multi-channel and synchronization cause serious degradation in data quality, posing challenges for accurate reconstruction. To address this issue, this paper proposes a data recovery algorithm based on matrix completion (MC) for 3D imaging of structured targets. The structural characteristics of architectural targets introduce a low-rank property into the data, stemming from the inherent correlation among adjacent pixels. Utilizing the principle of matrix completion, combined with the sparsity of scattering points in elevation, a low-rank and sparse joint completion model is established. Furthermore, the Truncated Schatten-p Norm and Sparse Regularizer-Alternating Direction Method of Multipliers (TSPN-ADMM) algorithm is adopted for solving. Additionally, considering that this recovery method reconstructs the 2D complex image, a Filter-MC processing framework is proposed to further enhance the performance. Finally, both simulation and real data verify the effectiveness of the proposed recovery method and framework.

Synthetic aperture radar automatic target recognition (SAR ATR) with limited data has gained significant attention as practical application requirements change. Despite many proposed methods, key problems caused by the limited SAR data remain under-researched, hindering further performance improvement. In this article, we establish an SAR ATR model based on causal theory. It compares the causal effect of SAR ATR between cases with ample and limited data, showing that the negative impact of the confounder, which is blocked with ample data, is introduced with limited data, resulting in poor performance of limited-data SAR ATR. To address this, we propose a limited-data SAR ATR causal method via dual invariance intervention, which first derives the causal interventional solution. This solution is transformed into two optimizable objectives: inner-class feature invariance and the independence of features from the confounder. Subsequently, the dual invariance mechanism is designed to filter SAR outlier samples and noise features under limited-data conditions, accurately obtaining the intraclass invariant feature. It also alleviates the need for ample SAR data when optimizing the independence of features from the confounder, achieving the two objectives. Finally, the proposed method not only unravels the key problem caused by limited data but also derives an effective solution with precise recognition performance. Extensive experiments on three benchmark datasets validate the rationality of the causal SAR ATR (CSA) model, the effectiveness of the solution, and the soundness and recognition performance of the method. The codes and more experimental results will be released athttps://github.com/cwwangSARATR/SARATR_Causal_Dual_Invariance.

Obtaining the clear contours of ship targets via Synthetic aperture radar (SAR) is extremely valuable for monitoring the sea. Now there are many deep learning imaging methods for ground scenes with good results, but they will face three main challenges when imaging ship targets: 1) The translational and rotational motions of ship targets during travel and due to waves respectively bring spatial invariant and variant errors that are tough to be estimated and compensated, resulting in defocused SAR imaging results; 2) The varying motion of ships demands high generalization ability of the imaging reconstruction network to adapt to the ship targets with sample-wise variant motion parameters; 3) Since ships are noncooperative targets, the accuracy of motion parameter estimation should be evaluated based on image quality, whereas the available SAR image quality assessment functions exhibit limited robustness. To address these issues, this article proposes a deep learning-based SAR imaging framework for ship targets via deep unfolding. Firstly, the motion model and characteristics of ship targets are analyzed, and the SAR imaging model for ship target with complex translational and rotational motion is established. Secondly, an imaging network with high generalization ability is proposed to adapt echoes for imaging under different motion parameters. On this basis, a SAR ship image quality assessment network is proposed to assess the imaging results of SAR ship targets with different focusing qualities. Then, the high-resolution imaging problem of ship targets is regarded as a motion parameter optimization problem, with the image quality assessment results as the objective function. Finally, this problem is optimized to search for the most accurate translational and rotational motion parameter variables of the ship target, achieving error compensation and imaging. The validity of the method is verified though the simulation of point targets and real SAR scenario data.

With multiple channels in azimuth receiving echoes, multichannel radar has the potential of forward-looking imaging, and various schemes can be formulated. However, due to the different resources utilized by different imaging schemes, the angular resolution will be different. How to analyze the angular resolution and then design appropriate parameters is a key issue in multichannel radar forward-looking imaging. In this paper, based on the echo model of forward-looking imaging, the imaging schemes of synthetic aperture and real aperture are illustrated firstly. Then, based on the ambiguity function theory, the angular ambiguity functions, and the analytical expressions of the angular resolution for different imaging schemes are derived and analyzed. Finally, the simulation results of point targets and extended targets are presented to verify the effectiveness of the theoretical analysis, which would lay a significant foundation for the design of forward-looking imaging schemes of multichannel radar.

Distributed synthetic aperture radar (SAR) tomography has recently aroused extensive attention due to its 3-D imaging capability and flexibility. However, atmospheric turbulence brings in motion errors, and collaborative work of multiple platforms increases the degree of freedom (DoF) for motion errors and introduces time-frequency synchronization errors. These undesired errors result in severe 2-D image defocusing, making SAR tomography 3-D imaging unattainable. To address these issues, a novel 3-D focusing scheme that can simultaneously compensate for high DoF motion errors and time-frequency synchronization errors is proposed. Firstly, the RCM offset induced by time synchronization errors is corrected through preprocessing. Then, image-quality-based 2-D autofocus is utilized to achieve 2-D focusing. Finally, the effects of linear phase and constant phase introduced by 2-D autofocus are eliminated by image registration and elevation autofocus based on maximizing image sharpness, respectively. Simulation results demonstrate the effectiveness of the proposed 3-D focusing scheme.

Conventional array radar systems require element spacing to be less than half a wavelength to mitigate grating lobes in forward-looking image, which results in a large number of elements and imposes a significant hardware burden. In this paper, an equivalent virtual uniform linear array is formulated with differential visualization technique and multiple snap-shot data. Then the regularized iterative adaptive approach (RIAA) is adopted to achieve superresolution imaging of the virtual linear array. Simulation results are given to illustrate the effectiveness of the proposed scheme.

To achieve high-quality forward-looking imaging of multi-channel radar, the fusion of synthetic aperture imaging result and real aperture superresolution result is usually necessary. However, the fusion processing by directly multiplying two results will seriously affect the image quality. In this paper, based on the analysis of the characteristics of synthetic aperture imaging results and super-resolution results, the logarithm transformation is introduced firstly on the synthetic aperture result before multiplication fusion processing to improve the image quality. Simulation results are illustrated to verify the effectiveness of the proposed scheme.

Microwave Photonic Synthetic Aperture Radar (MWP-SAR) holds significant promise for applications in Earth remote sensing owing to its exceptional imaging resolution. This radar technology emits ultra-wideband signals surpassing those of conventional radar systems. Hence, MWP-SAR exhibits the potential to generate pseudo-color images by exploiting scattering differences, thereby augmenting the information acquisition capabilities of MWP-SAR. This paper introduces a methodology for synthesizing pseudo-color images while preserving the high resolution of MWP-SAR. The proposed algorithm employs an optimization technique to identify subband echo channels exhibiting the most significant differences in scattering characteristics. Simultaneously, to safeguard the resolution of MWP-SAR, a fusion model is devised to integrate the full-resolution Synthetic Aperture Radar (SAR) image with the multi-subband image. Finally, a full-resolution pseudo-color image was successfully synthesized from the measured airborne MWP-SAR data.

Multichannel radar has the potential of forward-looking imaging, but its azimuth resolution is usually poor due to the restriction of the platform size. Many superresolution methods have been developed to improve its azimuth resolution and sparse Bayesian learning (SBL)-based methods are popular within them. However, traditional SBL methods suffer from the over-sparse problem for extended targets, and they always fails to achieve good performance when there are both point targets and extended targets. In this paper, by judging the types of targets to assign different weights for different targets, and then applying the weighted average to the update results of hyperparameters in SBL iterations, a modified SBL-based scheme of multichannel radar forward looking imaging is proposed, and simulation results are illustrated to verify its effectiveness.

Sea clutter suppression plays a crucial role in maritime moving target indication. However, in the bistatic SAR (BiSAR) system, traditional space-time adaptive processing (STAP) method can’t satisfy the expected performance due to severe range cell migration (RCM), Doppler frequency migration (DFM), nonstationary clutter, and spatio-temporal spectrum expansion caused by the internal motion of sea clutter. To issue these problems, a matrix sparse model based spatio-temporal spectrum recovery method is proposed. The proposed method mainly consists of three steps. Firstly, Generalized Keystone transform in preprocessing stage is used for RCM correction and DFM compensation. Then, multiple spatio-temporal samples acquisition strategy for cell under test is designed, to enhance the solution robustness. Next, a matrix sparse model for BiSAR spatio-temporal spectrum recovery is constructed and solved. Finally, the sea clutter suppression performance is verified with numerical simulations.

Synthetic aperture radar (SAR) usually uses linear frequency modulation (LFM) signal to image the target and clutter signal, which often causes the target to be overwhelmed by strong clutter. To solve this problem, we propose a waveform design method which can reduce the clutter and guarantee the SAR imaging performance. Firstly, the spectrum mask constraints of signal-to-clutter ratio (SCR) are obtained according to the scene prior information, and then the unimodular waveform of minimizing weighted integral sidelobe level (WISL) is designed. The final result is the SAR transmit waveform with spectral constraints. Simulation results show that the design can effectively achieve target enhancement and clutter suppression while maintaining SAR imaging performance.

High-resolution ship Inverse Synthetic Aperture Radar (ISAR) imaging is vital for identifying ship targets by revealing their contours and detailed features. However, the spatial-variant Doppler frequency caused by unknown motion parameters often leads to defocused ISAR images. Traditional Fourier-Transform-based methods struggle to handle the three-dimensional spatial-variant Doppler frequency (3-D SVDF). CLEAN-based methods, capable of estimating 3-D SVDF, often sacrifice target details. To overcome these hurdles, this paper proposes a refined ship ISAR imaging method based on Fractional Fourier Transform (FRFT). Employing FRFT, the method estimates multiple Doppler parameters and segregates scattering points into spatial-invariant sub-images based on their Doppler similarity. Phase compensation is then applied to each sub-image, enabling compensation for ship motion and extracting detailed ship target information in high-resolution ISAR images. Simulation results affirm the algorithm's efficiency in achieving image focusing and revealing intricate ship target details.

In the field of polarimetric synthetic aperture radar (PolSAR) automatic target classification (ATR), convolutional neural network (CNN) based methods have excelled owing to their adept feature extraction capabilities. However, these methods heavily rely on a sufficiently labeled training dataset for superior classification performance. Limited PolSAR training samples and inevitable noisy labels often render CNNs susceptible to overfitting. To tackle this challenge, a novel PolSAR image classification method employing beta mixture model and boundary amplification is proposed. Initially, the beta mixture model is utilized to fit the loss value distributions of noisy and clean samples, enabling the exploitation of distinct characteristics between these samples for probability estimation. Subsequently, to emphasize boundary samples, the boundary is delineated and expanded using the Sobel operator, amplifying losses for samples within this expanded region. Finally, a robust classification loss function is integrated into the training process to rectify losses incurred by network predictions. Experimental validation conducted on the Flevoland dataset demonstrates that the proposed method attains state-of-the-art performance.

Automatic target recognition (ATR) based on synthetic aperture radar (SAR) images has already obtained remarkable achievements on closed-set task. However, the recognition in a real-world scenario should not only identify the known classes but also appropriately deal with the unknown ones. To this end, we propose a SAR open-set recognition method aided by hierarchically reconstructive latent representation learning. First, a unsupervised representation learning via hierarchically-fused reconstruction network (HFRNet) is proposed to complement the lost information in supervised representation and obtain a preliminary closed-set result. Then, we adopt Openmax to correct closed-set recognition scores and give the probability of being the unknown ones, realizing the effective open-set recognition on SAR images. Finally, experimental results based on the measured dataset have shown the superior performance of the proposed method.

In recent years, iterative adaptive approach (IAA) has been proposed for super-resolution imaging in scanning radar, providing improved azimuth resolution. Traditional IAA involves computing the correlation matrix R for target scattering in each range cell, leading to iterative row-by-row solving and matrix inversion operations, causing high computational complexity. To this end, this paper proposes a Fast Batch-Based Iterative Adaptive Approach (FBB-IAA) that enables parallel and synchronized super-resolution processing of each range cell in the echo matrix. Additionally, it utilizes the two-dimensional conjugate gradient (2D-CG) method to avoid matrix inversion operation, significantly reducing the computational complexity compared to traditional IAA. Simulation results validate the superiority of the proposed method.

Target recognition is one of the most significant tasks in synthetic aperture radar (SAR) image interpretation. However, due to the varying difficulty in acquiring SAR images for different categories, SAR target recognition often encounters the issue of categories imbalance. This make majority categories contribute more to the loss than minority categories, yielding a decline in classification performance. To this end, a novel SAR target recognition approach under imbalanced categories is proposed. Firstly, focal loss (FL) is introduced to balance contributions of minority and majority categories to model optimization. Then, a first-order flatness constrained FL is devised to minimize the high generalization error effectively. Finally, a gradient norm aware minimization (GAM) algorithm is implemented to integrate first-order flatness into optimization process, yielding favorable recognition results for both minority and majority categories. Experimental results based on the moving and stationary target acquisition and recognition (MSTAR) dataset demonstrate the effectiveness of our proposed method.

Space-borne passive radar (PR) exploits non-cooperative signals of satellite constellations, specifically Low-Earth-Orbit (LEO) communication satellites. It offers advantages such as low intercept probability and reduced costs. However, utilizing LEO signals presents challenges as multiple similar signals from different satellites can interfere without identification. This paper proposes a method for LEO communication signal separation and target detection leveraging synchronization signal blocks (SSB) for discrimination and identification, generating a reference signal for subsequent processing to transfer target information to the range-Doppler domain. The final integration result enables the detection and localization of the target. The effectiveness of the proposed method is validated through simulated experimental results.

With the launch of high-resolution (HR) synthetic aperture radar (SAR) imaging satellites and the rapid development of convolutional neural networks (CNNs), ship recognition in HR SAR images has shown further improvements. Unlike ship targets in low and medium-resolution SAR images, which only possess a few pixels and present a spot-like appearance, ship targets in HR SAR images pose a larger area of pixels. However, CNN lacks the power to model dependencies between long-range features occupying large areas of pixels. A convolutional transformer (CvT) is introduced to deal with this issue. CvT integrates the local features of CNNs and the capability of capturing long-range dependencies of transformers to model both local and global dependencies for ship recognition in an efficient way. The cosine-margin loss is also applied to constraint strictly the distribution of the features to further improve the performance. Experimental results on the benchmark FUSAR-Ship dataset demonstrate the effectiveness of the proposed method for ship ATR in HR SAR images.

The Microwave Photonic (MWP) radar, designed for ultrawideband signal emission in high-precision imaging, encounters challenges in ultra-high-resolution images where strong scattering points obscure weak edge scattering areas of the target. Consequently, the resultant images exhibit isolated strong points instead of continuous edges on the physical structure, diminishing the interpretative capacity of highresolution radar images. This paper proposes a Microwave Photonic Inverse Synthetic Aperture Radar (MWP-ISAR) edge recovery method based on morphological theory to address this issue. The method employs image preprocessing, including dilation and edge detection, followed by the hough transform (HT) for accurate edge localization. Subsequently, An adaptive neighborhood enhancement operator is used to restore the target’s edge features. Notably, the method adeptly tackles the computational cost problem arising from numerous parameter estimates in the image reconstruction process based on parametric models while also automatically extracting edges for restoration. Finally, the proposed method undergoes validation and quantitative evaluation using authentic aircraft data, substantiating its effectiveness in enhancing radar image interpretation.

Multistatic synthetic aperture radar (MuSAR) has the capabilities of short-time high resolution imaging and multi-angle target observation. The imaging quality of coherent MuSAR depends on the distribution of wavenumber spectrum (WS). In this paper, an imaging performance improvement method based on WS trimming is proposed to obtain high-quality imaging results when the WS distribution is not ideal. First, the echo signals of MuSAR are derived. Then, the distribution of the WS is analyzed, and the WS trimming problem is transformed into a constrained multiple objective optimization problem (CMOP), which is optimally solved by multi-objective particle swarm optimization (MOPSO) algorithm. Finally, numerical simulation are performed to verify the effectiveness of the proposed method.

Synthetic Aperture Radar (SAR), with its all-weather and all-day operation, is an effective tool for earth observation. However, with the continuous intensification of global changes, current earth observation methods face challenges in meeting the demands for global coverage and timeliness. Moon-based SAR (MBSAR) has the advantage of long observation time and wide coverage. However, the imaging capability of MBSAR is limited by the orbit characteristics of the moon. Using the moon as the illumination source and the high-earth orbit (HEO) satellite as the receiving station (MH-BISAR) not only allows for a wide imaging area and flexible viewing angle, but also can effectively reduce the signal transmission power. This paper first establishes the motion model of MH-BISAR in a unified coordinate system, then calculates the basic conditions such as the required transmit power for MH-BISAR and compares it with MBSAR. Next, the imaging capabilities of MH-BISAR and imaging time of global areas within one month are analyzed. Finally, it is concluded that MH-BISAR has the advantages of high resolution, long observable time, large imaging coverage, and low system requirements for earth observation. It can serve as a powerful means for earth observation.

High-squint super-resolution imaging for airborne scanning radar is crucial in remote sensing and earth information observation. Various methods have been proposed to enhance the azimuth resolution of imaging. However traditional methods are often limited by the requirement for manual adjustment of hyperparameters. In this paper, we propose a hyperparameter-free high-squint super-resolution method for airborne scanning radar based on likelihood based estimation of sparse parameters (LIKES). Compared to traditional sparse imaging methods, our presented approach ensures super-resolution while addressing the issue of manual hyperparameter adjustment. Simulation results demonstrate the effectiveness of the proposed method.

Multichannel radar can resolve left/right ambiguity and has the potential of high-resolution forward-looking imaging. However, when only a limited number of channels are available, in order to have good performance of resolving left/right ambiguity, there is always grating lobe in the imaging results. In this paper, a scheme of grating lobe suppression based on self-supervised contrastive learning is proposed. Firstly, a generative adversarial network is employed to achieve the mapping from the source domain X (i.e. the imaging results with fewer channels) to the target domain Y (i.e. the imaging results with more channels). Then, a contrastive loss is introduced to ensure across-domain preservation of the content information like the distribution of ground objects, preventing the generator from making unnecessary changes in results when suppressing the grating lobe. At last, simulation results are given to illustrate the effectiveness of the proposed method.

Compared with the traditional tomographic synthetic aperture radar (TomoSAR) system, multistatic TomoSAR can overcome the physical size limitation, realizing high-resolution 3-D imaging via single pass. However, due to the minimum safety distance limitation between flight platforms, the maximum unambiguous imaging range of multistatic SAR is relatively small, which is difficult to meet the mapping needs of urban high-rise. To address this problem, this paper proposes a multistatic TomoSAR ambiguity suppression method based on multiple subbands. This method can effectively achieve ambiguity suppression and 3-D reconstruction of the target without grating lobes. First, a multistatic TomoSAR imaging model is established. Second, we analyze the mechanism of multiple subbands ambiguity suppression. Finally, we utilize the adaptivity of the sparsity Bayesian recovery via iterative minimum (SBRIM) algorithm to the number of targets to realize multistatic TomoSAR 3-D imaging. The effectiveness of the proposed method is validated by simulations.

Electromagnetic jamming can seriously affect the quality of synthetic aperture radar (SAR) images and pose significant obstacles to image interpretation. The additional degrees of freedom brought by multistatic SAR considerably contribute to the accurate extraction of jamming information. In this paper, a jamming detection and jammer localization method for the interrupted sampling repeater jamming (ISRJ) is proposed. First, the jamming components in multistatic SAR images are detected by utilizing the time delay characteristics of ISRJ. Then, based on the Doppler frequency invariance property of jamming, a system of equations for multiple receiving configurations is solved for jammer localization. The effectiveness of the proposed method is demonstrated through simulation.

Subaperture processing is a commonly used method for time-domain SAR echo simulation. However, due to the effect of geometry configuration for bistatic SAR, doppler aliasing will appear in the simulated echo data, and then there will be ambiguity in the imaging result. In this paper, by analyzing the causes of ambiguity and introducing the interpolation in range, a modified subaperture echo simulation method for bistatic SAR is proposed. At last, simulation results are illustrated to verify the effectiveness of the method.

In complex synthetic aperture radar (SAR) imaging configurations, such as bistatic SAR, time-domain algorithms are less constrained and more accurate than frequency-domain algorithms. But they have not be applied to real-time imaging well because of large computation. The back-projection algorithm based on wavenumber-domain spectral splicing (WFBP) that emerged recently can be used to resolve this contradiction. In this paper, an efficient implementation architecture of this algorithm is designed based on FPGA. Parallel structures are used for sub-aperture BP imaging and images fusion. The imaging results and speed of the system is verified by simulations and experiments. It runs WFBP over a image of size 1024×512 in 0.131s, significantly accelerating imaging while maintaining accuracy of time-domain algorithms.

In high sea states, ship targets often exhibit intense and complex rotational movements due to sea waves, leading to defocused and time-varying imaging results for bistatic synthetic aperture radar (SAR). This paper introduces a novel strategy for selecting optimal imaging times of ship targets by analyzing image projection plane (IPP). Utilizing short-time Fourier transform (STFT) on target echoes, rotation parameters of the ship are estimated via differential evolution (DE) method, to determine IPP throughout the entire observation period, so that the optimal imaging moments can be selected. By comparison of image contrasts and entropies, the optimal imaging time lengths are also refined. Overall, this paper offers an optimal time section strategy ensuring clear views of ship targets, and simulation results verify its effectiveness.

With the increasing widespread use of synthetic aperture radar(SAR) systems in various environments, jamming have become a serious problem that they face. In many situations, these jamming affect SAR systems’ ability to gather information. Many anti-jamming methods are based on the classification of jamming types. In order to provide information about jamming types, it is necessary to design a classification method which can classify multiple types of jamming. Considering the complexity of jamming features, in this paper, multi-domain jamming feature analysis and a residual network with convolutional block attention module (CBAM) are proposed to classify SAR jamming. To train this jamming classification network and validate its effectiveness, a database containing different jamming simulations is generated. The simulation results show that this method has reliable classification performance for various types of jamming.

Airborne multistatic synthetic aperture radar (SAR) shows various advantages such as high flexibility and target observation from different perspectives regarding localization. However, due to the separation of transmitters and receivers, the time-frequency synchronization error is induced and the traditional Range-Doppler (R-D) localization method is unreliable because of challenges in Doppler centroid estimation. Furthermore, there is a position error in the position of the unmanned aerial vehicle (UAV) as a high-maneuvering platform. This paper proposes a novel localization method for airborne multistatic SAR based on time difference of arrival (TDOA). This method only uses range information extracted from received signals without Doppler centroid estimation for TDOA localization equation construction. Then the localization equation is solved by the Taylor algorithm for the target location, which is of good robustness to the time-frequency synchronization error and the UAV position error. Numerical simulations validate the effectiveness of this method.

Deep learning has achieved remarkable success with the super-resolution of ordinary optical images. However, synthetic aperture radar (SAR) images have unique imaging mechanisms and features different from optical images, and are faced with problems such as low signal-to-noise ratio, limited resolution, speckle noise and sidelobe, which affect the readability and quality of images. Improving SAR image quality is an important research direction in SAR image processing, and the development of deep learning technology provides a new perspective for improving SAR image quality. Deep convolutional neural networks (CNNS) or other deep learning models are usually used for training and optimization, ignoring the multidimensional features of SAR images. Therefore, we propose a SAR image super-resolution reconstruction network based on multi-scale feature extraction. By considering the multi-dimensionality of SAR image features, the proposed algorithm achieves more accurate image reconstruction, and achieves good results in both quantitative and visual evaluation.

Real aperture radar (RAR) has a limited aperture of the antenna resulting in a low azimuthal resolution. To improve the azimuthal resolution, the L2regularization method is applied to the forward-looking scanning radar imaging. However, the traditional L2regularization method requires an inverse operation, which results in extremely low imaging efficiency. In this paper, we propose a fast frequency domain approach based on Gaussian prior for forward-looking scanning radar imaging. This method transforms the spatial domain inverse convolution problem to the frequency domain and avoids matrix inverse. The fast Fourier transform implementation greatly improves imaging efficiency. The simulation experiments demonstrate the effectiveness of the proposed method.

Super-resolution techniques based on the convolution model of target scattering coefficient and antenna pattem have been widely used in scanning radar forward-looking imaging for past few years. In previous work, the sparse regularization model is proved to be effective in solving the recovery problem of sparse scenes. However, traditional regularization methods usually only add constraints to the target and the regularization parameters are difficult to choose, thus easily causing noise amplification and image recovery distortion. In this paper, a parameter-free estimation method based on low-rank and sparse hybrid constraints is proposed. Firstly, based on the traditional regularization model, sparse and low-rank constraints are added to the target and background respectively to effectively suppress the noise amplification; then, a parameter-free estimation solver is proposed to solve the problem that the regularization parameter is much more difficult to choose. In addition, the superior performance of the proposed method is verified by simulations.

Radar forward-looking imaging is important in many fields, such as the ground mapping, autonomous driving. However, conventional monostatic SAR cannot realize forward-looking imaging due to left/right ambiguity. In this paper, based on the geometry model of multichannel forward-looking SAR, the principle of resolving left/right ambiguity is illustrated firstly, and then the performance is analyzed. At last, simulation results are illustrated to verify the effectiveness of theoretical analysis.

In recent years, the Sparse Iterative Covariance Estimation (SPICE) algorithm has been applied to Direction of Arrival (DOA) estimation in MIMO radar, significantly enhancing radar resolution and quality. However, the high computational complexity of this algorithm poses challenges for real-time processing performance. In this paper, we introduce an online implementation framework for MIMO radar DOA estimation based on the SPICE algorithm. By incorporating sliding window processing, the complexity of the parameter space in each iteration is reduced with minimal resolution loss. Compared to existing SPICE algorithms, the proposed online sliding window qSPICE method achieves substantial computational savings without sacrificing performance. Simulation results demonstrate the superior performance of the proposed method.

Sparse super-resolution algorithm has been used in scanning radar imaging to improve its azimuth resolution. For sparse targets, traditional super-resolution methods usually introduce L1norm to improve azimuth resolution. However, the results obtained based on the L1norm are usually biased estimates, which leads to the limited effect of improving the azimuth resolution. In this paper, a sparse target reconstruction method based on non-convex penalty term is proposed. On the one hand, in order to reduce the bias effect, the L1norm in the cost function is replaced with the SCAD (smoothly clipped absolute deviation) penalty term that is closer to the L0norm. On the other hand, ADMM method is used to solve multi-constraint problems, and we use iterative shrinkage threshold method to solve non-convex optimization subproblem. Compared with the traditional sparse super-resolution method, the proposed method has better performance. The superior performance of the proposed method is verified by simulation and measured data processing.

Multipass synthetic aperture radar (SAR) tomography is a promising method for 3-D imaging. However, atmospheric turbulence-induced motion errors lead to severe 2-D defocusing and hinder 3-D imaging. The spatial-variant characteristic of motion errors further complicates motion compensation. To address this issue, an improved autofocus method is proposed, which achieves 2-D focusing by reconstructing trajectory. Firstly, a spatial-variant phase error compensation model is established to elucidate the correlation between motion errors and 2-D image quality. Subsequently, to maximize image sharpness, motion errors estimation is performed using a combination of the block coordinate descent (BCD) algorithm and the gradient descent (GD) algorithm. Finally, image registration is applied to correct 2-D image offset induced by an image-quality-based optimization strategy. Simulation results confirm the effectiveness and superiority of the proposed method.

Video synthetic aperture radar (SAR) has attracted increasing attention in recent years due to its ability to provide continuous images for the scenes of interest. However, practical applications of video SAR are limited by the large amount of data and computational costs involved in imaging. In this paper, we propose a deep unfolding network for reconstructing SAR videos from undersampled echo data. Firstly, we introduce a low-rank representation operator to perform low-rank representation on the video SAR tensor. Then the problem of reconstructing video SAR is modeled as a regularization problem based on low-rank representation and solved by the alternating direction method of multipliers (ADMM) algorithm iteratively. Finally, we unfold the iterative solution into a deep neural network to learn the network parameters and low-rank representation operator from the data. Simulation experiments validate the effectiveness of the proposed method.

To enable the synthetic aperture radar (SAR) automatic target recognition (ATR) system to continuously adapt to new recognition scenarios, it is necessary to equip the system with the ability to quickly update models. However, when these models learn new tasks, the knowledge of old tasks is quickly forgotten, a phenomenon known as catastrophic forgetting. The reason for catastrophic forgetting is that the model does not use the features of old tasks sufficiently. In this letter, an exemplar-free class incremental learning based on maximizing mutual information (CIL-MMI) is proposed to solve this problem. To effectively use the extracted features, CIL-MMI actively clusters features to maximize the mutual information (MI) between features and corresponding labels. The proposed method successfully avoids the distribution overlap caused by the small interclass differences and large intraclass variances inherent in SAR images. Experiments on the Moving and Stationary Target Acquisition and Recognition (MSTAR) dataset indicate that the proposed method outperforms state-of-the-art approaches, demonstrating improvements of 5.41%, 1.93%, and 2.47% at incremental steps 1, 2, and 3, respectively.

This letter presents a novel method to resolve the ambiguity in forward-looking radar systems. Traditional methods for ambiguity suppression face challenges with antenna deviations and the ill-conditioning of measurement matrices. To address these limitations, we introduce a deviation-adaptive unfolding network (DAUNet), which integrates perturbed compressive sensing (PCS) and advanced deep learning techniques. The DAUNet efficiently handles the space variance of the measurement matrix by utilizing matrix direct summation and the Kronecker product and modeling the ambiguity suppression as a unified PCS problem. This method incorporates iterative learning processes and a novel neural network architecture to reconstruct ambiguity-free images from multichannel forward-looking radars. Simulation results demonstrate that our approach significantly outperforms the existing methods in handling both point targets and distributed scenarios.

Two-dimensional absolute velocity estimation of moving target is a key challenge for real-aperture scanning radar because 2-D velocity estimation methods suffer from low precision or heavy computation load. For example, the traditional Hough-transform-based method can only estimate the along-track velocity with one-order range migration. In this letter, a 2-D absolute velocity estimation method is proposed using the multiorder range migration information. The method extracts the multiorder range variation in adjacent echo sequences, and uses least-square linear fitting to estimate the along-track and cross-track velocity based on the relationship between the target echo and 2-D absolute velocity. Simulation experiments show that the velocity estimation error of the proposed method can be lower than 0.6 m/s, and the estimation time can be less than 1 s.

Synthetic aperture radar automatic target recognition (SAR ATR) methods often struggle due to inadequate training data. In this letter, we introduce a causal interventional ATR method (CIATR), specifically designed to address the challenges posed by limited synthetic aperture radar (SAR) data. This approach is key in revealing the underlying causal relationships among essential factors in ATR, enabling us to achieve the desired causal effect without altering the imaging conditions (ICs). To address the challenges in SAR ATR with limited data, we developed a structural causal model (SCM) based on causal inference principles. This model helps identify how ICs, as confounders, induce spurious correlations between SAR images and their classifications, which can be solved by standard backdoor adjustment. Our implementation of backdoor adjustment begins with data augmentation, employing a spatial-frequency domain hybrid transformation. This step is crucial in estimating the potential effects of varied ICs on SAR images. Following this, a feature discrimination strategy is introduced to incorporate a hybrid similarity measurement. This technique is essential for assessing and mitigating the impact of changing ICs on the features extracted from SAR images, focusing on both structural and vector angle influences. The CIATR method effectively uncovers the true causal relationships between SAR images and their classes, even with limited data. Tested on MSTAR and OpenSARship datasets, our method shows promising performance in limited data scenarios, achieving 75.05% for ten-way five shots.

Automatic target recognition (ATR) holds a crucial position in synthetic aperture radar (SAR) image interpretation. Despite deep learning advancements have significantly propelled SAR ATR, addressing the challenge of target recognition with a few training data remains a vital concern in SAR applications. Two main issues still exist: 1) In few-shot SAR ATR, the depth and width of CNN-based models are limited, which restricts its modeling capacity, and thus extracting discriminative generalized features remains challenging. 2) With only a few labeled SAR images, the resultant class distribution is biased due to the intra-class diversity and inter-class similarity of SAR samples, which degrades the recognition performance. To address these challenges, in this letter, we propose a novel dynamically weighted prototypical learning (DWPL) method. Firstly, to extract discriminative generalized features from SAR images, we propose a new convolutional transformer network with great capacity to capture long-range dependencies of local features, together with an effective random task augmentation strategy. Secondly, in consideration of intra-class diversity and inter-class similarity, a dynamically weighted prototypical module (DWPM) is designed to adaptively assign weights to the few labeled samples that have varying discriminative information. This enables the model to effectively explore the hidden features in few samples. Through experiments conducted on the moving and stationary target acquisition and recognition (MSTAR) dataset, our method achieves recognition accuracies of 97.22% and 92.01% for 3-way 5-shot and 3-way 1-shot SAR ATR tasks in SOC, revealing significant and robust recognition performance.

Synthetic aperture radar (SAR) is capable of obtaining the high-resolution 2-D image of the interested target scene, which enables advanced remote sensing and military applications, such as missile terminal guidance. In this article, the terminal trajectory planning for SAR imaging guidance is first investigated. It is found that the guidance performance of an attack platform is determined by the adopted terminal trajectory. Therefore, the aim of the terminal trajectory planning is to generate a set of feasible flight paths to guide the attack platform toward the target and meanwhile obtain the optimized SAR imaging performance for enhanced guidance precision. The trajectory planning is then modeled as a constrained multiobjective optimization problem given a high-dimensional search space, where the trajectory control and SAR imaging performance are comprehensively considered. By utilizing the temporal-order-dependent property of the trajectory planning problem, a chronological iterative search framework (CISF) is proposed. The problem is decomposed into a series of subproblems, where the search space, objective functions, and constraints are reformulated in chronological order. The difficulty of solving the trajectory planning problem is thus significantly alleviated. Then, the search strategy of CISF is devised to solve the subproblems successively. The optimization results of the preceding subproblem can be utilized as the initial input of the subsequent subproblems to enhance the convergence and search performance. Finally, a trajectory planning method is put forward based on CISF. Experimental studies demonstrate the effectiveness and superiority of the proposed CISF compared with the state-of-the-art multiobjective evolutionary methods. The proposed trajectory planning method can generate a set of feasible terminal trajectories with optimized mission performance.

With the echo difference between different channels, the left/right ambiguity in single-channel forward-looking synthetic aperture radar (SAR) can be resolved using dual-channel radar. However, due to the restricted channel resources, existing methods have limited performance in resolving left/right ambiguity, especially in the area along the flight path. In this article, a novel left/right ambiguity resolving scheme is proposed to improve the imaging performance of dual-channel forward-looking SAR. In the scheme, the steering matrix considering the time-frequency–space information of echo was designed first, and the linear mapping equation between the echo and the original scene was established. Then, the regularized iterative adaptive approach (RIAA) is introduced to solve the equation, thereby achieving dual-channel forward-looking SAR left/right ambiguity resolving and superresolution imaging simultaneously and directly in the echo domain. At last, the simulated and experimental results were illustrated to demonstrate the effectiveness of the proposed scheme.

Bistatic synthetic aperture radar (BiSAR) has high geometric diversity and can obtain the target information from different observation angles. With the ability of azimuth beam steering for both platforms, different bistatic imaging modes can be implemented to achieve better cooperation of beam footprints for enhanced imaging performance. For the multimode general bistatic SAR with complex trajectory (MGCT-BiSAR), due to the different beam steering methods and the translational-variant geometry, the spatial variance of the Doppler centroid becomes complicated, which leads to a severe azimuth spectrum aliasing problem. Moreover, the complex trajectory of MGCT-BiSAR may introduce high-order range cell migration (RCM) and Doppler parameters, which leads to the 2-D coupling and the different Doppler characteristics for different beam steering modes. The existing imaging algorithm cannot uniformly process the multimode SAR data. This article proposes an improved polar formatting algorithm based on minimum azimuth spectrum width (minASW-PFA) to uniformly process the bistatic SAR data. The proposed method first calculates the unified deramping factors and proposes a minASW bulk deramping method to minimize the azimuth spectrum width of the different imaging modes. However, the process leads to a severe range-Doppler coupling, which cannot be fully eliminated by the traditional polar formatting and wavefront curvature compensation methods. Therefore, an improved polar format mapping and wavefront curvature compensation method are also proposed to solve the coupling problem and achieve high-order spatial variance compensation. Finally, numerical simulations verify the proposed algorithm to achieve unified imaging for multimode general bistatic SAR with complex trajectories.

The features employed for synthetic aperture radar (SAR) target recognition have evolved from traditional SAR target geometric features and pattern features to modern deep features, indicating a trend of increasing recognition accuracy but decreasing feature interpretability. Therefore, the fusion of multidimensional features has been investigated by many researchers. Existing feature fusion methods typically involve simple concatenation or addition of geometric features and pattern features with deep features, or directly incorporating them into deep networks. However, such fusion methods mentioned above inadequately consider the potential conflicts between features and hard to fully exploit multidimensional features. To solve the above problem, a multidimensional feature joint learning framework (MFJL-Framework) that serves the SAR target recognition task is proposed in this article, which consists of three models. Specifically, the SGC-GA-Model can select pattern features for SAR targets based on geometric feature constraints, the Global and local Feature Information interaction Capture model (GFIC-Model) can select deep features with high-level abstract semantics, and the MFFS-Model can complement and fuse these two types of features to maximize the utilization of feature information. Experiments and comprehensive ablation studies on four datasets, namely OpenSARShip-1.0, FUSAR-Ship, MSTAR-T72Variants, and SAR-AIRcraft-1.0, collectively demonstrate that the recognition performance of our proposed FJL-Framework outperforms the current state-of-the-art methods.

Synthetic aperture radar (SAR) target recognition plays an indispensable role in interpreting SAR images. However, differences in radar parameters (including factors such as imaging modes and imaging angles) often lead to resolution differences between training and test data, posing challenges for existing methods in recognizing SAR targets under cross-resolution conditions. To address this issue, this article proposes a domain adaptation (DA) framework based on feature aggregation and compensation (FAC) for cross-resolution target recognition in SAR imagery. Initially, we employ a unique local vision transformer (LocalViT) to establish global and local adversarial networks that capture invariant features under cross-resolution conditions. Following this, we design a multiscale feature fusion module (MSFFM) to capture multiscale semantic features of targets at different resolutions. Subsequently, we propose a novel class feature aggregation module (CFAM) to map targets of varying resolutions onto the unit sphere, thereby aggregating features of samples from the same class and distinguishing those of samples from different classes. Finally, we narrow down the difference in frequency-domain information of targets at different resolutions by developing a resolution semantic compensation module (RSCM). This module compensates for the semantic feature about resolution during target recognition across varying resolutions by converting high- and low-frequency information. The experimental results on three SAR datasets (OpenSARShip, FUSAR-Ship, and SRSDD-v1.0) confirm that our method outperforms the state-of-the-art (SOTA) DA methods, with an increase of 2.06%, 1.96%, and 1.61% in three sets of cross-resolution scenes, respectively.

Deep neural networks have shown remarkable effectiveness in SAR target recognition. However, the explainability problem for deep neural networks remains insufficiently addressed. One approach to tackle this challenge is the SHAP method. It enhances the explainability of deep neural networks in SAR target recognition by observing how the target, shadow, and clutter regions play their own distinct roles. The masked regions are typically filled with Zero, Mean, or Random values in optical images. But if the same operation performed on SAR images, it will affect the distribution of clutter and thus introducing new out-of-distribution challenge. In this paper, we propose a novel masking method to enhance the reliability and efficiency of the SHAP method in SAR-ATR applications. Experimental results on the MSTAR and OpenSARShip-1.0 datasets demonstrate that our proposed method provides a more faithful representation to show the importance of every single regions in SAR target recognition. Compared to methods using Zero values, Mean values, and Random baselines, our proposed method significantly enhances the reliability of explainability.

High-resolution synthetic aperture radar (HR-SAR) is extensively used in ground remote sensing applications, including disaster monitoring and resource crop assessment. This is attributed to its exceptional high-resolution imaging capabilities. Microwave photonics technology plays a crucial role in enhancing the performance of SAR systems. It enables the direct emission of ultra-wideband signals across multiple frequency bands. However, this advancement also makes microwave photonic synthetic aperture radar (MWP-SAR) susceptible to complex and diverse electromagnetic interference. Particularly, it is vulnerable to radio frequency interference (RFI), and this interference seriously affects the high-resolution imaging results. In order to solve these problems, an MWP-SAR imaging algorithm based on depth spectral sensing and spectral reconstruction that is suitable for complex electromagnetic environments is proposed. Simultaneously, a sensing-removing-recovering (SRR) anti-interference imaging framework is established. To address the high-precision detection of various unknown interferences in MWP-SAR echoes, a deep learning network is constructed. This network is based on spectrum sensing theory and achieves an interference detection probability greater than 98% in various interference scenarios. Subsequently, targeting the continuous spectrum loss in the MWP-SAR signal after interference removal, a signal reconstruction algorithm is proposed. This algorithm employs Toeplitz transformation for signal loss scenarios, facilitating the recovery of signals with extensive spectrum loss post-interference removal. The factor group sparse regularization (FGSR) algorithm is used to quickly solve the signal recovery problem. Through simulation and measured data processing, the superiority of the proposed algorithm over existing interference suppression imaging algorithms is demonstrated.

Due to its all-day and all-weather capability, synthetic aperture radar (SAR) plays an important role in many remote sensing and monitoring applications. However, conventional SAR image reconstruction methods generally perform undifferentiated imaging, complicating target detection. To address this challenge, we propose a SAR image reconstruction method based on self-attention deep prior learning for differentiated image reconstruction and target detection. The proposed method can separate the target from the clutter using their feature priors during image reconstruction, thus helping improve target detection performance. Specifically, a deep prior learning operation based on a self-attention convolutional neural network (SACNN) is proposed. SACNN can help enhance target and suppress clutter by learning both the local and global features. Finally, the proposed method is implemented through an unrolled deep network with a loss function designed to make the reconstructed image beneficial for target detection. Simulation experiments have been conducted to verify the efficacy of the proposed method.

Sea clutter suppression plays a crucial role in maritime moving target indication. However, in the bistatic SAR (BiSAR) system, traditional space-time adaptive processing (STAP) method can’t satisfy the expected performance due to severe range cell migration (RCM), Doppler frequency migration (DFM), nonstationary clutter, and spatio-temporal spectrum expansion caused by the internal motion of sea clutter. STAP based on sparse recovery (SR-STAP) is an effective method for clutter suppression, but two major problems still remain. (1) The multiple samples for solution need to satisfy the same spatio-temporal distribution characteristics. Nevertheless, such consistency is not applicable when considering violent internal motion of sea clutter. (2) The computational complexity is exceedingly high. To issue these problems, an efficient matrix sparse recovery STAP (MSR-STAP) method based on Kronecker transform is proposed. The proposed method mainly consists of three steps: (1) Generalized Keystone transform in preprocessing stage is used for RCM correction and DFM compensation. (2) Multiple spatio-temporal samples acquisition strategy for CUT is designed, to enhance the solution robustness. (3) An efficient MSR-STAP model is established and solved. Subsequently, the space-time filter is designed without clutter covariance matrix estimation, to facilitate effective sea clutter suppression. Compared with existing SR-STAP methods, computational complexity of the proposed method decreases by orders of magnitude, and the spatio-temporal spectrum expansion effect is greatly reduced. The sea clutter suppression performance is verified with numerical simulations.

Localization and tracking are important components for ground-moving target indication (GMTI). The traditional range-Doppler (RD) localization model is always applied to locate stationary targets in bistatic synthetic aperture radar (BiSAR). However, for moving targets, this localization model is no longer applicable due to the strong coupling of position and velocity with Doppler frequency. To address the severe challenge, this article proposes a joint moving target localization and tracking method for BiSAR-GMTI via transmitter–receiver trajectories extraction and inversion. The key innovation is the derivation of closed-form localization results for moving targets, which is beneficial for the quantitative analysis of localization error. The proposed method is structured into three main parts. First, to accurately capture the range history trajectory information, a segmented fitting extraction and inversion framework is designed. Second, to estimate the moving target’s initial state and demonstrate localization observability, joint range localization equations are established where the closed-form localization result of the moving target is derived. Finally, by defining the appropriate state transition equation and observation equation, and incorporating particle filtering (PF), the position measurement errors of bistatic platforms are weakened, and accurate tracking of moving targets is realized. The proposed method only utilizes range information, mitigating localization errors caused by Doppler estimation inaccuracies effectively. Simulation experiments and real data both demonstrate the localization and tracking accuracy of moving targets.

Radar forward-looking imaging has important applications in autonomous landing, autonomous navigation, reconnaissance guidance and other fields. However, conventional single channel synthetic aperture radar (SAR) or Doppler beam sharpening (DBS) technology has a blind area for forward-looking imaging due to left/right ambiguity and small angle variation. Multichannel radar can utilize the differences of echoes from multiple channels in azimuth to resolve left/right ambiguity, and has the potential for forward-looking imaging. However, there is still a problem of low azimuth resolution due to the restriction of array size. In this article, a deep learning based multichannel radar forward-looking super-resolution imaging framework is proposed. In this framework, synthetic aperture processing is conducted on the echo data of each channel to obtain the image with left/right ambiguity, and the preliminary forward-looking imaging is achieved first by resolving left/right ambiguity with multichannel data. Then, the generative adversarial network (GAN)-like network with mixed attention mechanism is designed to learn the mapping relationship between the original scene and the preliminary imaging result. At last, based on the learned mapping relationship, the echo data of multichannel radar is processed with the proposed framework to achieve forward-looking superresolution imaging. Experimental results were provided to verify the effectiveness of this imaging framework.

This article presents a robust and efficient method for recovering missing data in sub-Nyquist synthetic aperture radar (SAR) using matrix completion (MC) techniques. Unlike previous MC-based methods that have limitations on the required data properties, such as broadside-looking mode, sparse target scenarios, distributed missing, or low missing data ratios (MDRs), our method is capable of handling consecutively clustered missing data and high MDR, even in the presence of weakened spectrum sparsity caused by squint-looking mode or dense target scenarios. Our approach involves arranging the incomplete SAR echo data from a single range-time sampling instant into a Hankel structured matrix and recovering it using nuclear-norm-based convex relaxation and an augmented Lagrangian multiplier (ALM) method. To address computational complexity, we propose segmenting the SAR echo data into smaller slices and providing guidelines for selecting an appropriate slice size. We demonstrate the effectiveness and accuracy of our method through extensive experiments using simulated data and realistic Radarsat-1 data. The experiments cover various challenging conditions, including squint-looking mode, dense target scenarios, wide-swath coverage, high MDR, and clustered missing. The results highlight the capabilities of our method in handling sub-Nyquist SAR data recovery, even in challenging scenarios. Our proposed method offers a robust and efficient solution for recovering missing SAR data.

Forward-looking radar (FLR) has been widely discussed because of its super-resolution capability. However, for the high-speed radar platform, the super-resolution performance of FLR degrades significantly due to the limited signal model accuracy. In this article, to observe the azimuth–elevation information of multiple targets based on a high-speed radar platform, a sparse Doppler phase convolution (SDPC) model is proposed by randomly and sparsely scanning the radar beam to reduce the coherent processing interval (CPI) and limit the signal model errors. On the one hand, the Doppler phase is introduced to characterize the vector superposition relations of the echo in each azimuth–elevation direction, thus limiting the error of the conventional convolution model (CM). On the other hand, an azimuth–elevation sparse scanning scheme is proposed to reduce the CPI, allowing for accurate second-order approximation of the range history and further limiting the reconstructed errors for high-speed radar platforms. In addition, the velocity application boundary and the sparsity boundary of the SDPC model are quantitatively analyzed. Simulations compare and validate the performance of the proposed SDPC model with the conventional CM using three classical super-resolution algorithms. Based on the proposed model, azimuth–elevation information of multiple targets can be accurately reconstructed on high-speed radar platforms.

Deconvolution methods can be applied in airborne scanning radar to enhance its angular resolution for improving the collision avoidance ability in the forward-looking direction. However, as the movement speed of the airborne platform increases, the traditional convolution signal model cannot be applied because of the model errors in the amplitude profile and Doppler phase. In this article, an accurate profile-phase deconvolution method is proposed to achieve scanning radar forward-looking super-resolution imaging, particularly for high-speed platforms. On one hand, a profile-phase convolution (PPC) model is established by analyzing the influence of high-speed platform on echo amplitude profile and Doppler phase. The proposed model accurately captures the variation of beam dwell time caused by the coupling of platform motion and beam scanning, which directly affects the echo amplitude profile. On the other hand, relying on the proposed PPC model, an adaptive regularization (AR) deconvolution method is derived to avoid hyperparameter selection. Point-target and surface-target results demonstrate that the proposed PPC model and the AR deconvolution method are competent for super-resolution imaging on high-speed platforms.

Frequency modulation continuous wave (FMCW) radar has been paid much attention in forward-looking navigation applications because of its no-blind-range capability. However, after dechirp processing, pulse interference signals may appear in the range time domain, which seriously pollutes the whole radiation direction. In this article, a cross-domain low-rank and sparse (CD-LRS) optimization framework is proposed to enhance the angular resolution and suppress the pulse interference signals based on the scanning mode of its antenna. On the one hand, to cut off and recover the polluted signals, a low-rank spectra reconstruction approach is proposed by utilizing the low-rank characteristic of the Hankel matrix formed by the interference-rejected data in the range time domain. On the other hand, to suppress the residual interference signal and enhance the angular resolution simultaneously, an adaptive sparse reconstruction method is formed in the azimuthal time domain by adopting an alternating direction method of multipliers (ADMMs)-based solver. Compared with the traditional anti-interference methods, the proposed framework can enhance the angular resolution and suppress the interference signals based on the signal features in different domains. Simulations and experimental results are applied to verify the effectiveness of the proposed framework.

Swarm unmanned aerial vehicle synthetic aperture radar (UAV SAR) system leverages multiple UAVs to form a formation, overcoming the limitations of a single platform and enabling the execution of advanced SAR missions. By forming a uniform linear array formation, the swarm UAV SAR system is able to coherently enhance resolution in one direction. Extend to 2-D cases, a uniform planar array needs to be formed for resolution enhancement. However, the requirement for a large number of UAVs to form the planar array can lead to significant costs. In addition, the performance of resolution enhancement is intricately tied to the chosen system formation. Therefore, there is a pressing need to conduct research on methods to obtain the optimal formation. In this article, a hybrid resolution enhancement (HRE) framework has been proposed for the swarm UAV SAR system to optimize resolution enhancement performance while mitigating costs. The proposed framework is mainly divided into two stages: cost-effective formation strategy and optimal HRE. The cost-effective formation strategy, which lays down the foundation for resolution enhancement, is comprised of three steps. First, to achieve HRE with a reduced number of UAVs, a cross-shape formation structure is established. Second, to effectively optimize the position and velocity of the central node of the UAV swarm for optimal resolution performance, a constrained differential evolution (DE)-nondominated sorting (CDE-NS) algorithm is proposed. Third, baseline design is conducted to determine the attached nodes’ positions for optimal coherent resolution enhancement (CRE). After the ideal formation is obtained, optimal HRE can be accomplished. Specifically, the principle of CRE is explained. The inspiration, motivation, and novelty of the proposed noncoherent resolution enhancement method named minimum combination (MC) are elucidated. Simulation results have demonstrated the validation of the proposed framework.

Bistatic synthetic aperture radar (BiSAR) parametric imaging can reconstruct the structural information of the target, which is one of the research hotspot for BiSAR imaging. However, compared with monostatic SAR, there are several challenging problems to be faced in terms of BiSAR parametric imaging, such as complex echo models, compute and storage burden, azimuth-dependent phase (ADP). To this end, we first propose the BiSAR parametric echo model, followed by analysis of the characteristics of BiSAR parametric imaging. Based on the abovementioned anaysis, we propose a deep adaptive BiSAR parametric imaging network (DAPI-net), which consists of three parts, namely adaptive dimensionality reduction module, parameter estimation module and image reconstruction module. The main idea of DAPI-net is to employ coarse imaging to determine the approximate range of target parameters and construct a local observation matrix to reduce computational complexity. On this basis, a variant observation matrix learned solver and ADP compensation unit are used to achieve high-precision BiSAR target parameter estimation. Among them, the ADP compensation unit mainly mitigates echo model mismatch caused by ADP and improves the parameter estimation performance. Finally, BiSAR parametric image output is achieved through image dilation. The efficacy of DAPI-net is validated through simulation experiments and microwave anechoic chamber data. The proposed algorithm not only addresses the identified challenges but also advances the state-of-the-art in BiSAR parametric imaging.

Geosynchronous synthetic aperture radar (GEO-SAR) can provide long-duration and wide beam coverage over the interested target scene, which is an ideal illuminator for bistatic SAR acquisitions. As a particular system configuration, the GEO bistatic SAR with maneuvering platform as the receiver (GEO-MP-BiSAR) can achieve continuous observation of the interested target during the flight. The target recognition and tracking information can be generated from the updating images for enhanced guidance performance. However, the bistatic SAR imaging performance is dependent on the observation geometry, which in turn is determined by the trajectory of the receiver. Therefore, in this paper, the trajectory optimization for GEO-MP-BiSAR is firstly investigated. The goal of the method is to generate a set of feasible trajectories to guide the maneuvering platform towards the target, and meanwhile obtaining the optimized imaging performance during the whole flight. The trajectory optimization is then modeled as a multi-objective optimization problem with multiple constraints, where the trajectory control and SAR imaging performance are comprehensively considered. Then, a knee-guided multiobjective evolutionary algorithm is put forward to effectively solve the problem, where the knee solutions are utilized to guide the search process and improves convergence and diversity of the method. The proposed algorithm can generate the prescribed number of solutions with significant trade-offs between the performance metrics. The mission designer can then choose a solution from only a few optimized candidates for implementation, which greatly improves the efficiency of decision making. Experimental studies demonstrate the effectiveness of the proposed method.

A conventional synthetic aperture radar (SAR) transmits the linear frequency modulation (LFM) signal for imaging and processes the target and clutter without distinction, making target enhancement difficult to realize. Waveform design is considered a common approach to improve the signal-to-clutter ratio (SCR). However, due to the undulating frequency response of radar clutter, the waveform designed for SCR improvement tends to have a high sidelobe autocorrelation function (ACF), which can lead to the degradation of SAR imaging performance. To solve this problem, this article proposes a complementary SAR waveform set design method for the improvement of the SCR while suppressing the sidelobe. Different from the traditional single waveform design method, in this article, the transmitted waveform set is jointly optimized. The designed waveform with a complementary sidelobe varies between pulses. Utilizing the multipulse azimuth compression of SAR imaging, the aforementioned high sidelobe can be eliminated after azimuth focusing. To this end, we use the inexact alternating direction penalty method (IADPM) and develop the spectrally constrained complementary sequences (SCCSs) algorithm to solve the resulting nonconvex optimization problem. Simulation and experimental data verification highlight the effectiveness of the proposed design for SCR improvement while maintaining the SAR imaging performances.

Synthetic Aperture Radar (SAR) is widely used in various fields due to its all-weather and all-day working characteristics. With the increasing use of SAR on small platforms, SAR is facing a series of problems due to the large volume of echo data. Imaging methods based on compressed sensing (CS) use the sparsity prior of the scene to reconstruct images from undersampled echoes. However, the CS-based method requires the imaging scene or its transformation domain to be sparse, which is not the case for most practical applications. This paper proposes a deep unrolling network named NSR-NET, which is based on SAR image representation learning and is applicable for undersampled imaging in non-sparse scenes. In modeling, the learned image representation is adopted as the regularization term. Then, the proximal gradient descent (PGD) algorithm was used to derive the iterative solution of the model. In network design, the iterative process is unrolled into a deep neural network with learnable parameters. Specifically, image representation is obtained through 2D convolutional layers in the network, and a learnable piecewise linear layer is used to fit the regularization function, which ultimately achieves the mapping from undersampled echoes to SAR images. Comparative experiment using different imaging methods shows that the imaging performance of the proposed network exceeds that of the state-of-the-art methods in non-sparse scenes. Moreover, we also designed transferability validation experiments with different radar parameters and imaging scenes, whose experimental results suggest that the proposed network has good generalization ability.

Sidelobe suppression is an important step in SAR imaging, especially for forward looking areas with poor azimuth resolution. In this paper, the signal model of multichannel forward-looking SAR imaging is realized with BP algorithm firstly, and then spatial smoothing coherence factor(SSCF) is introduced to suppress sidelobes of the imaging results. At last, simulation results are given to illustrate the effectiveness of the proposed scheme.

With multiple channels receiving echoes in azimuth, multichannel radar has the potential of forward looking high resolution imaging. However, due to the limitation of platform size, its azimuth resolution is poor. In this paper, by considering the effect of platform motion and off-grid problem, a multichannel radar forward looking superresolution imaging method based on atomic norm minimization is proposed. Simulation results are illustrated to verify the effectiveness of the method.

Different sea states make maritime ship targets move three-dimensionally, causing imaging results severely defocused. Therefore, selection of proper imaging time has attracted global attention in the field of ISAR imaging. This paper proposes an imaging time and plane selection algorithm based on short-time fractional Fourier transform (STFrFT). By STFrFT, time-frequency curves of the target echo for different fractional orders are obtained, and probability density functions of frequency distributions are offered to estimate when Doppler frequency of the target changes most stably. Then, different imaging planes of different STFrFT orders are analyzed to select the optimal imaging time moment. Finally, by comparing image contrasts of various imaging lengths, the optimal imaging time is selected. In general, this method can select the optimal imaging moment, imaging length and imaging plane, and simulation verifies the effectiveness of this imaging method.

Currently, regularization methods are widely applied to radar forward-looking super-resolution imaging, but imaging performance is greatly affected by the regularization parameter. In order to address this issue, the regularization method with weak-depended on parameter for forward-looking super-resolution imaging is proposed in our work. First, the objective function is established under the premise of sparse target prior; then, iteratively reweighted solver is applied to resolve the objective function. The key idea is to join the regularization weighting factor in the process of solving sparse regularization problem, which reduces the sensitivity to the regularization parameter and avoids the imaging error caused by unreasonable parameter selection. Compared to traditional sparse regularization method, the proposed method is less dependent on the regularization parameter, and the imaging performance is superior under the same conditions. Simulation results verify the effectiveness of the proposed method.

The actual detection scenes faced by bistatic synthetic aperture radar (SAR) often have elevation information, which will cause shadow and overlay effects in imaging results. In view of the above problems, this paper proposes an echo generation method of bistatic SAR based on hidden point removal(HPR) operator. In this paper, according to the bistatic configuration, the shadow area is deduced first by using blanking algorithm(HPR operator). On this basis, the visibility of the target in the scene can be determined to obtain the echo. Finally, a simulation using a digital elevation model is conducted to verify the accuracy of the proposed method.

In UWB (ultra-wideband) radar imaging, there are differences in the signal scattering characteristics of targets for different frequencies. The estimation of target scattering characteristics is an important research field in target recognition and structured imaging. To solve this problem, this paper combines the target scattering characteristic estimation and super-resolution imaging into a low-rank and sparse signal reconstruction problem. It proposes an ADMM (the alternating direction algorithm of multiplier) scattering characteristic estimation algorithm based on low-rank and sparse signal. First, the echo model of UWB signal is established according to GTD model. Then build the corresponding complete dictionary. Finally, the ADMM-LS algorithm is used to solve the multi-objective joint optimization problem. Numerical simulation experiments verify the algorithm to prove the effectiveness of the proposed method.

Conventional monostatic SAR or DBS technology cannot realize forward-looking imaging due to Doppler symmetry ambiguity. With multiple channels in azimuth, multichannel radar can resolve the left/right ambiguity, and have the potential for forward looking imaging. However, multichannel radar have flexible channel configuration, and there are different processing schemes, which may have different imaging performance and challenges. In this papar, based on the signal model of SIMO radar, different processing schemes for multichannel radar forward-looking imaging are illustrated and simulated, as well as the corresponding potential and challenges.

Multichannel radar has the potential of forward-looking imaging, but its azimuth resolution is usually poor due to the restriction of the platform size. Many superresolution methods have been developed to improve its azimuth resolution. However, these methods always have the problem of grid mismatch. In this paper, sparse Bayesian learning (SBL)-based multichannel radar forward-looking super-resolution imaging considering off-grid error is proposed, and simulation results are illustrated to verify its effectiveness.

In the past, the parameters of SAR system were relatively fixed, which lead to the poor flexibility in the use of imaging resources. In this paper, the concept of cognitive radar is introduced into SAR, which can make the system adaptively optimize the resources required. We design a three-steps process for large-scale imaging, which consists of two reconnaissance process and one cognitive processing link. Finally, we design the parameters and resources of two reconnaissance process manually or by using genetic algorithm. Experimental results show that the parameters designed meet the demand of large-scale search and the genetic algorithm used can significantly reduce the resource cost in further reconnaissance and improve the flexibility of system.

Moving target indication (MTI) plays an important role in both military and civilian applications. However, the performance of MTI has deteriorated dramatically due to clutter interference. In bistatic synthetic aperture radar (BiSAR), MTI mainly faces three challenges: First, BiSAR has large range cell migration (RCM); Second, the Doppler spectrum is severely extended; Third, the clutter presents nonhomogeneity and nonstationarity. In order to solve these problems, a novel MTI method for single channel BiSAR is proposed. At first, the first-order keystone transform (KT) is performed to remove linear range walk (RWK) regardless of unknown motion parameters. Then, the high-order RCMC is realized by reference compensation function. It is extremely difficult to achieve MTI because moving target is submerged in clutter whether in azimuth time domain or Doppler domain. So, to achieve separation of moving target from stationary clutter effectively, a novel procedure named increased dimension rotation (IDR) processing is introduced. Thereafter, the distribution characteristic of moving target and stationary clutter after rotation processing is analyzed exhaustively. Naturally, the optimal filter is constructed for clutter suppression in the light of the distribution characteristic. Finally, the energy of moving target can be accumulated after inverse increased dimension rotation and Fractional Fourier Transform (FrFT). And the MTI in BiSAR can be easily achieved.

The information obtained by single sensor is limited, so the fusion of multi-source images becomes the research focus. Monostatic SAR and bistatic SAR can obtain complementary scattering information from the same target, but the SAR image lacks color information resulting in limited visual effects. The optical image contains color information but it is difficult to achieve high resolution at the decimeter level due to the limitations of present optical cameras. Therefore, this paper proposes an algorithm to fuse monostatic and bistatic SAR image with optical image. First, the three images are registered, then the monostatic and bistatic images are stitched into a dual-channel image. Finally, the dual-channel SAR image is fused with optical image using a blind model-based fusion method. The fusion result combines the color information of optical image with the high-resolution and texture information of SAR image, which enhances the visual effect and readability, providing great convenience for subsequent applications. The fusion results of the experimental data show that the algorithm can effectively make the image information more comprehensive and accurate.

Recently, the iterative adaptive approach (IAA) has been widely applied to enhance the azimuth resolution of real beam mapping (RBM) imagery. However, the IAA suffers from extremely high computational complexity in practice. This paper proposes a Split IAA for sparse scene to reduce the complexity. First, the IAA cost function is decomposed. Then the echo data is split into blocks, and the iterative model is redefined according to the target block and its corresponding cost function. Consequently, the high-dimensional data inversion problem is decomposed into multiple low-dimensional sub-problems to achieve fast super-resolution imaging. The measured results show that the proposed Split IAA significantly reduces the computational complexity without affecting super-resolution performance.

The joint FPGA and multi-DSP imaging architecture has found widespread use in airborne SAR and satellite-based SAR. However, traditional back-projection (BP) algorithms are not efficient enough for real-time imaging. This paper proposes a back-projection algorithm based on wavenumber-domain spectral splicing (WFBP) for designing an efficient SAR imaging system. A unified polar coordinate system is employed in this work to project the sub-aperture imaging results, which eliminates the need for a large amount of interpolation and multiple projection transformations. Furthermore, spectral shifting is applied to eliminate the effects caused by wavenumber-domain spectral overlap. Experimental results demonstrate that the WFBP algorithm reduces the imaging time by 64% compared to the traditional BP algorithm.

Recently, a two-dimensional (2-D) scanning radar super-resolution model has been proposed to simultaneously achieve azimuth-pitch super-resolution imaging. However, due to the addition of the pitch dimension, the complexity of the state-of-art methods becomes extremely high. In this paper, based on the sparse learning via iterative minimization (SLIM), we propose a low-complexity 2-D sparse scanning radar super-resolution method. First, the signal model of 2-D scanning radar is established. Then, base on the traditional SLIM method, the 2-D scattering estimation of the target can be iteratively solved by exploiting the conjugate gradient (CG) algorithm and the Kronecker product property. Compared with the existing methods, the proposed method has lower computational complexity and stronger adaptive ability without losing resolution performance. The simulation verifies the effectiveness of the proposed method.

Synthetic aperture radar (SAR) target recognition networks performance has been remarkably improved, posing serious exposure risks to our high-value targets. Researches have shown that it is valid to protect our high-value targets by generating adversarial examples. However, most existing SAR adversarial examples generation approaches are based on the premise that irregularly global perturbation data can be directly added to SAR images, which is difficult to implement in practice. To this end, a target partial-occlusion SAR adversarial examples generation approach is proposed in this paper. First, the target region in SAR image is extracted using the combination of OTSU algorithm and morphology operations. Then, the random search (RS) algorithm is introduced to optimize the occlusion position in the extracted target region with the constraint of occlusion area and value, so as to misclassify the SAR target recognition networks. Experimental results based on the moving and stationary target acquisition and recognition (MSTAR) dataset have shown the effectiveness of the proposed method.

Multistatic SAR based on UAV swarm can reduce synthetic aperture time and increase the imaging frame rate while maintaining high image resolusion, thus it is of great promising prospect. Due to the lack of efficient and high precision imaging methods, the performance of UAV swarm SAR is not fully developed. Therefore the imaging method suitable for UAV swarm SAR is studied in this paper. Compared with frequency domain imaging techniques, time domain imaging method is more appropriate for UAV swarm SAR with complex trajectories. But the heavy computational burden is a problem. Thus this paper proposes a fast imaging method based on spectrum fusion for high frame rate UAV swarm SAR. In the method, each bistatic pair in UAV swarm SAR is split into several sub-pairs to generate coarse angular resolution images and the fusion of images is realized through spectrum fusion, which obtains the imaging result of UAV swarm SAR. Experiment is carried out to verify the effectiveness of the proposed method.

Most of the existing angular super-resolution techniques are based on the convolution model of the target scattering coefficient and the antenna pattern, and this convolution relationship provides the potential for improving the angular resolution. However, due to the non-ideal working environment in practical applications, the antenna pattern generates phase and amplitude errors, resulting in model mismatch. Model mismatch produces errors during super-resolution processing, degrading the final imaging quality. In order to address the issue, an angular super-resolution method of real aperture radar under model mismatch condition is presented in this paper. First, we introduce an error matrix on the original convolutional model to consider model mismatch errors. Secondly, the target sparse prior is exploited to construct the objective function under the model mismatch condition. Finally, the alternating direction method of multipliers (ADMM) solver is utilized to solve the objective function, obtaining the final imaging result. Simulations verify the effectiveness of the proposed method.

To design configuration parameters for coherent multistatic synthetic aperture radar (C-MuSAR), a wavenumber spectra projection (WSP) approach is proposed in this paper based on the relationship between the wavenumber support regions (WSRs) and configuration parameters, including synthetic aperture time, positions and flight directions of receivers. First, the projected pattern of multiple WSRs is deduced, and the relationship between multiple WSRs and the point spread function (PSF) is analyzed. Second, the primary WSR is designed based on the relationship between the transmitter and the leading receiver. A WSP method is proposed to quickly deduce the configuration parameters of the following receivers. Finally, based on the designed configuration parameters of C-MuSAR, an adaptive WSP method is adopted to reconstruct the targets. Simulations are carried out to testify the proposed method.

Marine surveillance radar plays an important role in marine environment monitoring, however, its detection performance is often affected by sea clutter. In this paper, we consider the sea clutter suppression process as the mapping from clutter radar data domain to clutter-free radar data domain, and propose a new sea clutter suppression method based on clutter cancellation generative adversarial network (CCGAN). The proposed CCGAN contains sea clutter suppression generator (SCSG) and clutter-free domain discriminator (CFDD). With the proposed network, the clutter suppression result can be obtained. To ensure the target imformation is not affected while sea clutter is suppressed, the proposed method introduces target consistency loss in addition to adversarial loss during the training process. Experimental results have shown the proposed method can achieve excellent clutter suppression performance.

Multistatic synthetic aperture radar (SAR) can obtain abundant information from different angles for terrain classification and tomography. However, multistatic SAR systems, particularly the multistatic global navigation satellite systems (GNSS) face the problem of insufficient resolution. To address this issue, a novel noncoherent combination method termed as minimum combination (MC) is proposed. Inspired by dual apodization, MC is specifically designed to improve the resolution of the multistatic SAR system. Relative to the conventional noncoherent addition (NA) method, MC yields an image with higher resolution and reduced sidelobe level. Simulation results are presented to illustrate the effectiveness and superiority of the proposed approach.

In the field of maritime surveillance, ship recognition in synthetic aperture radar (SAR) images is a significant and fundamental step. Compared with traditional methods, convolutional neural networks (CNNs) tend to be the mainstream in SAR ship recognition. However, these methods ignore one core issue. Multi-scale features can enhance the expression ability of features, which are currently not well-exploited. In response to this problem, a novel feature weaving pyramid network (FWPN-Net) is proposed. FWPN-Net contains a multi-scale feature weaving module (MFWM), which can integrate high level semantic information and low level detailed information to obtain better representations of multi-scale SAR ship features. Experimental results on benchmark dataset OpenSARShip show that the proposed FWPN-Net performs better than classic CNN methods and modern SAR ship recognition CNN method.

Recently, deep learning has gained increasing popularity in array signal processing. In this paper, we estimate the direction of arrival (DOA) for the multiple-input and multiple-output (MIMO) radar system based on deep learning. First, we convert DOA estimation into a linear inverse problem with spatial sparsity, and construct a neural network based on the iterative shrinkage thresholding algorithm (ISTA) to improve the interpretability of the network. Then, a stacked denoising autoencoder (DAE) is employed to achieve data-driven denoising, which improves the anti-jamming ability of DOA estimation. Finally, a new deep unfolded network named denoising learned ISTA (Denoising-LISTA) is proposed for DOA estimation. Simulation results illustrate that the proposed method improves the robustness of DOA estimation with single snapshot sampling and keeps significant predominance in beam sharpening and sidelobe suppression.

Scanning radar (SR) super-resolution imaging has been widely reported recently. However, most of the existing methods are based on the amplitude convolution relationship between the reflectivity function and the antenna pattern, realizing super-resolution imaging through deconvolution. When the radar platform moves fast, Doppler phase and pattern distortion caused by the movement destroys the above convolution relation and decreases the super-resolution performance. In this paper, we proposed a pattern distorted complex convolution model to describe SR super-resolution imaging under high-speed platform. Simulations show that the proposed model is more suitable for SR super-resolution imaging under high speed platform.

Multichannel SAR can solve the problem of left/right ambiguity and has the potential of high-resolution forward-looking imaging. However, its azimuth superresolution in the adjacent area of flight path is still poor due to the small angle variation. In this paper, we propose a multichannel forward-looking SAR azimuth superresolution method based on CycleGAN. First, the multichannel forward-looking SAR imaging results are obtained, and then the mapping between forward-looking imaging results and the ground truth is learned with CycleGAN, to achieve azimuth superresolution. At last, simulation results are given to illustrate the effectiveness of the proposed method.

In the problem of ship detection in complex scenes, in addition to the characteristics of ship targets, there is rich semantic information in the global and local background of the whole scenes, which provides more valuable inference information for ship detection. Therefore, in this paper, we propose a ship detection method in complex scenes considering both global and local information perception for SAR images. Firstly, the proposed method detects the globally stable region and the locally significant region respectively, and then designs a judgment method combining the two to eliminate false alarms, so as to ensure that the detected target has both globally stable characteristics and locally significant characteristics. The detection performance of the proposed method is verified by the spaceborne SAR images covering the coastal areas. The result shows that the proposed method can effectively detect ships in complex scenes, especially eliminating most false alarms in land areas.

Recently, multistatic airborne SAR (MuA-SAR) is becoming a research hotspot due to its flexibility. Multi-platform data fusion requires that the imaging algorithm has strong adaptability to the flight path and relative spatial configuration of the airborne platforms. Therefore, the time domain algorithm based on back projection (BP) is suitable. However, in the existing BP-based methods, data needs to be projected into each grid one by one. In fact, not all pixels are target pixels that need to be projected, and the back projection of non-target pixels leads to a lot of invalid computation. Applying these methods directly to MuA-SAR will inevitably lead to a great increase in computation. To reduce the redundant back projection operation of BP algorithm and improve the efficiency of imaging processing in MuA-SAR, a fast imaging method based on segmentation before imaging is proposed in this paper. On the basis of fast factorized back projection (FFBP) algorithm architecture, an image segmentation method based on maximally stable extremal regions (MSER) is introduced. In the process of recursive fusion at each stage, only the pixel information of the segmented suspected target area is transferred to the next stage for fusion, and then the imaging efficiency is improved. The simulation and comparative experiments verify the effectiveness of the proposed method.

Ground-to-air real aperture radar scans the airspace to acquire range-azimuth-pitch 3D echoes for imaging. In real situations, the echo data may be corrupted by interference, making it difficult to reconstruct the original scene using echoes with missing data. To overcome this problem, a fast superresolution imaging method under the non-uniform sampling model is proposed in this paper. Firstly, a non-uniform sampling model is proposed to well model the echoes with missing data, which makes it possible to reconstruct the real target distribution from the echoes with missing data. Secondly, we use the sparse regularization (SR) super-resolution imaging method to reconstruct the real target distribution. Since the high dimension of the dictionary matrix leads to the expensive computational cost of the SR method, we propose a fast superresolution imaging algorithm based on low-rank approximation to reconstruct the targets quickly. Simulation results show that the original scene can be effectively reconstructed from the echoes with missing data based on our proposed model, and the proposed algorithm greatly improves the computational efficiency compared with the traditional methods while not leading to a loss of imaging performance.

Ground-based two-dimensional (2D) array radar suffers from low angular resolution, including azimuth and pitch directions because of the limited size of antenna aperture. In this paper, to improve the two-dimensional angular resolution for a ground-based 2D array radar, a 2D low-rank Truncated Singular Value Decomposition (2D-LRTSVD) superresolution algorithm is proposed by transforming the 2D deconvolution problem into several low-rank inversion problem. First, the traditional 2D convolution signal model is transformed as a low-dimensional signal model by dividing the Kronecker product matrix into several low-dimensional steering matrices. Second, a 2D-LRTSVD method is proposed by compressing the data dimensions of the low-rank steering matrices. Based on the proposed method, the operational complexity can be reduced by avoiding direct high-dimensional matrix inversion. Finally, the 2D angular resolution of array radar can be enhanced without performance loss in a low computational complexity. Simulations are carried out to verify the proposed method.

Several radar systems have been proposed in the past decades, including real aperture radar (RAR) and synthetic aperture radar (SAR). Spatial resolutions of different radar systems cannot be compared together because their work modes are different. In this paper, a normalized spatial resolution analysis model is proposed to deduce the spatial resolution of different systems. First, the normalized wavenumber spectra of different radar systems are deduced. Second, the relationship between spatial resolution and the wavenumber spectra distribution is analyzed. Finally, the point spread functions (PSFs) of different radar systems are simulated.

Recently, super-resolution techniques have been widely used in real aperture radar super-resolution imaging. And the majorize-minimization(MM) algorithm was recently introduced for scanning radar applications, resulting in substantial improvements in the angular resolution and quality of the processed images. Regrettably, the computational complexity and storage cost are high and quickly increase with growing data size, limiting the applicability of the estimator. In this paper, we strive to alleviate this problem, deriving an online MM algorithm, allowing for efficiently updating of the sparse reconstruction result for each online radar measurement along the scanned beam. The proposed method is a regularized extension of the current MM implementation, which not only offers constant computational and storage cost, independent of the data size, but also provides enhanced robustness over the current MM algorithm. Our experimental assessment, conducted using simulated data, demonstrates the advantage of the online MM(OMM) algorithm in the task of sparse reconstruction for scanning radar.

Reasonable analysis of radar interference effect is of great significance for adjusting jamming strategy in radar counter-measures (RCM). The modern battlefield is confronted with non-cooperative targets, so the conventional offline evaluation methods are difficult to apply. In this paper, a comprehensive evaluation method for radar interference effect based on the integrated cloud model is proposed. Firstly, a multi-layer index system for interference effect evaluation is established. Subsequently, the entropy method is employed to determine the weight of each indicator. To avoid the occurrence of hypertrophy as an imaginary number, the cloud parameters for each indicator are calculated using a modified inverse cloud generator. Eventually, a comprehensive assessment of the interference effect can be obtained by drawing the integrated cloud. The experimental results show that the proposed method is effective and can be applied to the evaluation of interference effectiveness in non-cooperative environments.

Imaging and target detection are usually regarded as two independent parts in conventional processing, which means that the detection performance will be affected by the imaging result. In this paper, the LRSD-ADMM-net is proposed to achieve simultaneous super-resolution imaging and target detection for forward-looking scanning radar. In addition, simulation results were provided to verify the effectiveness of the proposed algorithm.

Recently, a sparse super-resolution method relying on L1iterative reweighted norm (IRN) has been proposed to improve the imaging resolution of scanning radar. However, the method has poor adaptability due to the noise-sensitive user-parameter. To this end, an adaptive L1iterative reweighted sparse super-resolution method with no user-parameter is derived. Firstly, the scanning radar super-resolution model is established. Secondly, the user-parameter selection in the L1-IRN method is analyzed. Finally, the adaptive iteration weights are derived by transforming the sparse estimation problem into a maximum posterior (MAP) estimation problem. Compared with the existing L1-IRN method, the proposed method does not have any user-parameter, so it has adaptability to different signal-to-noise ratios (SNR) and is more robust. Simulation verifies the superiority of the proposed method.

Multichannel radar has the potential of forward-looking imaging with multiple channels receiving echoes on a single platform, but its azimuth resolution is usually poor. Superresolution algorithms have been developed to solve the problem, however, most of the methods have problems of difficulty in parameter adjustment and large amount of computation. In this paper, driven by the powerful learning ability of deep networks, the conventional ISTA reconstruction process is mapped into a deep network, and then the deep unfolding ISTA-Net is formed and used to realize multichannel radar forward-looking superresolution imaging. The simulation reults verify that the proposed method can provide high-quality reconstruction results while substantially reducing the imaging time.

Synthetic aperture radar (SAR) imaging for moving target can obtain complete situational awareness information of the detection area, and can realize the monitoring and control for moving target in the region of interest, which has important military and civilian dual-use value. However, due to the complex motion of target, the processing results of the existing SAR imaging methods severly defocused. In this paper, a high resolution SAR imaging method for moving target is proposed. First, we eliminate the coupling induced by linear range cell migration (RCM) by keystone transform. Then, the particle swarm optimization algorithm (PSO) is utilized to estimate the Doppler frequency rate, which can solve the problem of Doppler frequency rate mismatching when azimuth compression. Simulation results verifies the effectiveness of the proposed method.

In recent years, many countries are actively deploying Low-Earth-Orbit (LEO) communication satellite constellations, which have the advantages of both high power flux density (PFD) on the surface of the earth and large signal bandwidth. From the perspective of radar application, these new LEO constellations are very suitable as opportunity of illuminator for target detection in passive radar systems. In this paper, the echo signal using LEO communication satellite is analyzed, and a moving target detection method is proposed.

To obtain high-resolution images of the objects in front of platform, a super-resolution scheme for multichannel radar forward-looking imaging considering failure channels and motion error is proposed in this study. In the scheme, a failure channel detection method based on the correlation of pulse-compressed data of different channels is presented first, and then a revised steering matrix considering failure channels and motion error is constructed. Finally, the echo data are processed by the iterative adaptive approach (IAA) with the revised steering matrix. Simulation results are given to illustrate the effectiveness of the proposed scheme when dealing with failure channels and motion error.

Conventional monostatic SAR or DBS technology cannot realize forward-looking imaging due to Doppler symmetry ambiguity. Although the ambiguity can be resolved by using multiple channels in azimuth, its azimuth resolution in the vicinity of flight path is usually poor due to the small angle variation. To solve the above problems, a hybrid real/synthetic aperture scheme for multichannel radar forward-looking imaging is proposed in this paper. In the scheme, the synthetic aperture imaging result with left/right ambiguity is obtained firstly using the information of platform motion. Then the real aperture superresolution imaging with regularized iterative adaptive approach(RIAA) is achieved using the instantaneous data of multiple channels. At last, the two imaging results are fused to obtain the forward-looking superresolution image without left/right ambiguity. Simulation results are given to illustrate the effectiveness of the proposed scheme.

Shadow simulation is a key part of SAR raw data and image simulation for undulated terrain, which will directly determine the accuracy of SAR simulation and will also affect the accuracy of subsequent shadow-based SAR image processing applications. In this paper, a shadow simulation scheme for undulated terrain SAR images based on facet cell fitting and elevation angle comparison is proposed. In the scheme, the height information of each grid point on the same line of sight within the beam irradiation range is obtained firstly with the facet cell fitting, and the elevation angle of each grid point can be computed and compared to determine whether the occlusion happens. Then, all grid points in the illumination range are traversed to obtain the shadow judgment result at a certain azimuth moment, and the SAR shadows are obtained by superimposing the judgment results at all azimuth moments. At last, simulated and measured data experiments are given to verify the effectiveness of the proposed scheme, and the results show that this scheme can accurately simulate the SAR shadow of undulating terrain.

Retrieval of building height from synthetic aperture radar (SAR) image is of great significance for damage assessment after natural disasters, urban development and planning, and dynamic time series monitoring in urban areas. Based on the prior information of the building and the radar platform, the building height can be estimated by iteratively simulating the SAR image and matching it with the measured SAR image. However, the estimation accuracy would be affected inevitably when there is error of prior information introduced by instable platform or inaccuracy of measuring instruments. In this letter, the beam incident angle and building height are both used as search variables for iterative simulation and matching (ISM), and normalized mutual information (NMI) is used to measure the similarity between the simulated image and the measured image to achieve an accurate estimation of building height. At last, measured data experiments are provided to verify the effectiveness of the proposed scheme.

Multiview high-resolution range profiles (HRRPs) of aerial targets contain more target information than single-view one and will benefit accurate classification. However, feature information in HRRPs dynamically varies across different views, thus a dynamic classification framework is required to adjust the structure of the network along with the feature information variations and effectively make full use of those multiview features. To this end, we propose a dynamic-structure neural network (DSNN) with skip extraction and adaptive fusion blocks to adjust the network structure and adaptively fuse multiview features, enabling accurate aerial target HRRPs classification. In the skip extraction block, the skip gate automatically changes the block depth of each view to fit feature information variations, which ensures multiview HRRP features are dynamically exploited and extracted by the network. Then, in the adaptive fusion block, features from different views are weighted by the adaptive weight gate and effectively fused using associated attention, which further contributes to the classification. Besides, since the skip gate dynamically downsizes the extraction block for some views, the computational cost of DSNN is also reduced to some extent. Experimental results demonstrate that the proposed method has superior aerial target multiview HRRPs classification performance and computational efficiency over other state-of-the-art methods.

Recently, the sparse ℓ1-alternating direction method of multipliers (ADMM) based estimation algorithm was introduced for scanning radar, resulting in significant enhancements of the azimuth-pitch angular resolution. Regrettably, not only this method selects the hyperparameters manually, but also its time and space complexity increases rapidly with the data size, which restricts the capacity for applying in hardware system. To this end, a 2D weighted sparse iterative algorithm is derived in this paper, allowing for the hyperparameter-free and efficient sparse reconstruction result of scanning radar for the simultaneously azimuth-pitch resolution enhancement. The proposed method is a fast 2D extension of the current weighted Sparse Iterative Covariance-based Estimation (WSPICE) algorithm, which not only offers much less computational and storage cost, but also enjoys the adaptability with no hyperparameter and finer resolution. Experimental results of simulation and measured data demonstrate the advantage of the proposed method in azimuth-pitch resolution enhancement for scanning radar.

Sparse superresolution algorithms have been applied in scanning radar imaging to improve its azimuth resolution. However, the inverse matrix in each iteration is usually diagonal loading by the updating result, which leads to huge computational complexity for two-dimensional echo data. In this letter, a batch-processing superresolution framework is proposed to process the echo data in parallel. On the one hand, the optimization problem for sparse target recovery is modified as matrix form, which presents batch-processing potential for two-dimensional echo data. On the other hand, the optimization problem is solved by the proposed alternating direction method of multipliers (ADMM)-based batch-processing framework, which can avoid high-dimensional matrix inversion along different range bins. Compared with traditional sparse superresolution methods, the proposed batch-processing framework is much suitable for two-dimensional echo data superresolution.

Existing synthetic aperture radar automatic target recognition (SAR ATR) methods have been effective for the classification of seen target classes. However, it is more meaningful and challenging to distinguish the unseen target classes, i.e., open set recognition (OSR) problem, which is an urgent problem for the practical SAR ATR. The key solution of OSR is to effectively establish the exclusiveness of feature distribution of known classes. In this letter, we propose an entropy-awareness meta-learning method that improves the exclusiveness of feature distribution of known classes which means our method is effective for not only classifying the seen classes but also encountering the unseen other classes. Through meta-learning tasks, the proposed method learns to construct a feature space of the dynamic-assigned known classes. This feature space is required by the tasks to reject all other classes not belonging to the known classes. At the same time, the proposed entropy-awareness loss helps the model to enhance the feature space with effective and robust discrimination between the known and unknown classes. Therefore, our method can construct a dynamic feature space with discrimination between the known and unknown classes to simultaneously classify the dynamic-assigned known classes and reject the unknown classes. Experiments conducted on the moving and stationary target acquisition and recognition (MSTAR) dataset have shown the effectiveness of our method for SAR OSR.

Maritime surveillance is indispensable for civilian fields, including national maritime safeguarding, channel monitoring, and so on, in which synthetic aperture radar (SAR) ship target recognition is a crucial research field. The core problem to realizing accurate SAR ship target recognition is the large inner-class variance and inter-class overlap of SAR ship features, which limits the recognition performance. Most existing methods plainly extract multi-scale features of the network and utilize equally each feature scale in the classification stage. However, the shallow multi-scale features are not discriminative enough, and each scale feature is not equally effective for recognition. These factors lead to the limitation of recognition performance. Therefore, we proposed a SAR ship recognition method via multi-scale feature attention and adaptive-weighted classifier to enhance features in each scale, and adaptively choose the effective feature scale for accurate recognition. We first construct an in-network feature pyramid to extract multi-scale features from SAR ship images. Then, the multi-scale feature attention can extract and enhance the principal components from the multi-scale features with more inner-class compactness and inter-class separability. Finally, the adaptive weighted classifier chooses the effective feature scales in the feature pyramid to achieve the final precise recognition. Through experiments and comparisons under OpenSARship data set, the proposed method is validated to achieve state-of-the-art performance for SAR ship recognition.

Highly constrained multiobjective optimization problems (HCMOPs) refer to constrained multiobjective optimization problems (CMOPs) with complex constraints and small feasible regions, which are commonly encountered in many real-world applications. Current constraint-handling techniques will face two difficulties when dealing with HCMOPs: 1) feasible solution is hard to be found and too much search effort is spent in locating the feasible region and 2) since the total feasible region of an HCMOP can consist of several disconnected subregions, the search process might be stuck in the comparatively larger feasible subregion, which does not contain the whole Pareto front (PF). To address these two issues, an evolutionary algorithm with constraint relaxation strategy based on differential evolution algorithm, that is, CRS-DE, is proposed in this article. In each generation, the CRS-DE relaxes the constraints by dividing the infeasible solutions into two subpopulations based on total constraint violation, that is, the "semifeasible" subpopulation (SF) and "infeasible" subpopulation (IF), respectively. The SF provides information on the promising regions of finding the feasible solution and is the driving force for convergence toward the PF, while the IF focuses on global exploration for new promising regions. Corresponding reproduction and selection strategies are devised for the SF, IF, and feasible subpopulations, which create a clear division of labor with cooperation to facilitate the search for feasible solutions. To leverage the influence of CRS and prevent the population from premature convergence, a mobility restriction mechanism is developed to restrict the individuals in the SF and IF from entering the feasible subpopulation and enhance the diversity of the whole population. Comprehensive experiments on a series of benchmark test problems and a real-world CMOP demonstrate the competitiveness of our method compared with other representative algorithms in terms of effectiveness and reliability in finding a set of well-distributed optimal solutions for HCMOPs.

In our earlier study, an energy-efficient passive UAV radar imaging system was formulated, which comprehensively analyzed the system performance. In this article, based on the evaluator set, a mission planning framework for the underlying energy-efficient passive UAV radar imaging system is proposed to achieve optimized mission performance for a given remote sensing task. First, the mission planning problem is defined in the context of the proposed synthetic aperture radar (SAR) system and a general framework is outlined, including mission specification, illuminator selection, and path planning. It is found that the performance of the system is highly dependent upon the flight path adopted by the UAV platform in a 3-D terrain environment, which offers the potential of optimizing the mission performance by adjusting the UAV path. Then, the path planning problem is modeled as a single-objective optimization problem with multiple constraints. Path planning can be divided into two substages based on different mission orientations and low mutual correlation. Based on this property, a path planning method, called substage division collaborative search (Sub-DiCoS), is proposed. The problem is divided into two subproblems with the corresponding decision space and subpopulation, which significantly relax the constraints for each subproblem and facilitates the search for feasible solutions. Then, differential evolution and the whole-stage best guidance technique are devised to cooperatively lead the subpopulations to search for the best solution. Finally, simulations are presented to demonstrate the effectiveness of the proposed Sub-DiCoS method. The result of the mission planning method can be used to guide the UAV platform to safely travel through a 3-D rough terrain in an energy-efficient manner and achieve optimized SAR imaging and communication performance during the flight.

As high-frame-rate synthetic aperture radar (SAR) has the ability to form continuous SAR images and dynamically monitor the ground areas of interest, it has attracted more and more attention nowadays. In practical applications, the enormous data in high-frame-rate SAR system to obtain the multiframe images brings big challenges to its transmission, storage, and processing. In order to solve this problem, there are many recent papers on formulating the high-frame-rate SAR imaging problem into a low-rank tensor recovery problem, and correspondingly, the sampling amount of the high-frame-rate SAR data can be largely reduced. However, existing algorithms to solve the low-rank tensor recovery problem in high-frame-rate SAR application still suffer from large computational cost. Under the above inspiration, this article proposes a deep neural network architecture for high-frame-rate SAR imaging, i.e., tensor alternating direction method of multiplier network (TADMM-Net), which is more computationally efficient in the imaging procedure. Specifically, we formulate the high-frame-rate SAR imaging processing into a low-tubal-rank tensor recovery problem. We solve the low-tubal-rank tensor recovery problem using a tensor alternating direction method of multiplier (ADMM) algorithm and then design a new deep neural network architecture by applying algorithm unfolding techniques to the underlying low-rank tensor recovery problem. The proposed TADMM-Net approach shifts the computational burden from the testing phase to the training phase, and the practical processing time can be extremely decreased compared with the existing algorithms for the low-rank tensor recovery problem in high-frame-rate SAR imaging applications. It also offers various advantages over the existing high-frame-rate SAR imaging algorithms, including higher performance in the case with low sampling amount, lower storage complexity, and no requirement for handcraft hyperparameters adjustment. The methodology was tested on high-frame-rate SAR data. These tests show that the proposed architecture outperforms other state-of-the-art methods in high-frame-rate SAR imaging applications.

In the beam steering bistatic synthetic aperture radar (BS-BiSAR) system, both the transmitter and receiver beams can be steered in the azimuth direction, offering increased flexibility in mission planning to meet specific imaging requirements. Compared with the current SAR system, the system resources of BS-BiSAR have a high degree of freedom (DOF), including bistatic configuration, beam steering modes, and pulse repetition frequency (PRF), which are closely related to imaging performance. With appropriate beam steering strategy and observation geometry, better cooperation of beam footprints of the platforms can be achieved for enhanced imaging performance. The aim of this paper is to explore the resource management problem (RMP) for BS-BiSAR and optimize the comprehensive system performance, including imaging area, spatial resolution, signal ambiguity and radiometric performance. The RMP is then formulated as a single-objective problem subject to multiple constraints. An improved particle swarm optimization method combined with a sentry learning strategy is proposed to solve the optimization problem under strict constraints and high dimensional solution space. Simulation experiments are conducted to validate the effectiveness of the proposed method for managing system resources and optimizing performance. The result of the resource management method can be applied to guide the system design and achieve the comprehensive performance optimization of BS-BiSAR.

Compared to conventional microwave regime radars, microwave photonic (MWP) radar is capable of transmitting extremely large bandwidth signals, wherein the frequencies of such signals distribute across multiple bands. In practical applications, the large bandwidth of MWP radar may be split into multiple discrete sub-bands due to various considerations such as anti-jamming, resource-saving, communication band avoidance, etc. Nonetheless, it leads to the fact that MWP radar suffers from the challenging problems of side-lobes elevation and main-lobes broadening. These problems will affect the image quality seriously. In order to address this issue, a spectrum recovery algorithm based on an improved Truncated Schatten-pNorm and Sparse Regularizer-Alternating Direction Method of Multipliers (TSPN-ADMM) network is proposed in this paper. This algorithm can efficiently recover the lost spectrum in MWP radar applications and further improve the imaging quality of the MWP radar. In the lost spectrum recovery problem, the parameters of the recovery algorithm directly determine the recovery performance. The different forms of lost spectrum possessed by MWP radar make the selection of parameters for the spectrum recovery algorithm extremely difficult. As a consequence, in this paper, the spectrum recovery problem for MWP radar can be reformulated into a matrix completion problem by exploiting its joint sparsity and low-rankness. Based on the traditional TSPN-ADMM algorithm, an improved TSPN-ADMM-Net approach is proposed by utilizing the algorithm unrolling technique, wherein the hyperparameters in TSPN-ADMM algorithm are optimized in an end-to-end training manner. Consequently, the algorithm proposed in this paper can achieve excellent recovery results when dealing with the multiple spectrum missing situations existing in MWP radar. The effectiveness of the algorithm is verified by a combination of numerical simulations and actual MWP radar data.

Synthetic aperture radar (SAR) plays an important role in remote sensing by providing electromagnetic images of the observation scene. The prior knowledge-based SAR image reconstruction method can reduce the requirement of data sampling ratio and improve image quality. The existing prior knowledge-based method usually uses the magnitude information of SAR images while ignoring the phase information. However, since the echo and backscattering coefficients are complex values, the phase information of SAR images will help improve the reconstruction accuracy in the image reconstruction process. To improve the reconstruction performance, this paper proposes a SAR image reconstruction and autofocus method using complex-valued feature prior. In the proposed method, the complex-valued feature prior is learned from data by a complex-valued feature projection operator (CFPO), which can characterize and extract scene features in Range-Doppler domain and 2D frequency domain. The proposed CFPO enables more efficient use of echo data and helps to improve image reconstruction and autofocus performance. In addition, the proposed method is implemented by an unfolded deep network, which enables data-driven feature learning and efficient computation. The proposed method is verified by simulated and measured data.

The traditional Synthetic Aperture Radar Automatic Target Recognition (SAR/ATR) algorithm can train a sufficient number of known class samples and classify the samples in the test set. However, if the old model is trained only with the new class samples, the old class samples’ knowledge is easily forgotten by the new model, which is called catastrophic forgetting. The reason is that the model only fits the distribution of current training samples, so training the whole data set is necessary. Due to the limitation of storage resources, it is often not feasible to retain the whole data set. In order to avoid this phenomenon, a small number of old class samples can be kept to train with the new class samples. Therefore, how to select the old class samples becomes the key point. In this paper, the Density Coverage-Based Exemplar Selection (DCBES) is proposed to choose the key samples of the old class. DCBES selects samples based on the metric learning theory and the set covering theory. First, the metric learning theory is used to measure the similarity between samples and to obtain the density range of samples. Then the exemplar selection problem is considered a set covering problem, to select a fixed number of exemplars to achieve the maximum coverage of the class density range. Experiments on the Moving and Stationary Target Acquisition and Recognition (MSTAR) dataset show that our method is superior to other exemplar selection methods and achieves the best results.

Despite its widespread use in Earth remote sensing, synthetic aperture radar (SAR) image reconstruction remains challenging. The difficulties mainly lie in the handling of diverse scenes and motion errors with sparsely sampled data. Existing matched filtering (MF)-based methods cannot handle sparsely sampled data, while regularization-based methods lack adaptability to scene diversity. Although deep learning-based SAR methods can deal with these two issues, their performance will be degraded by motion errors. To address this, we propose a Transformer-based SAR image reconstruction method called RATIR-Net. The proposed method can obtain SAR images of various scenes under sparse sampling and motion errors by learning the correlations between echo data. In RATIR-Net, CNN-based encoding and decoding blocks are constructed to implement azimuth processes of range profiles (RP) in the range-Doppler domain according to the MF-based method. Meanwhile, a Residual Attention Transformer (RAT) block is designed to extract correlations between RPs, compensating for information loss caused by sparse sampling and suppressing non-correlated perturbations caused by motion errors. The CNN-based encoding and decoding blocks help reduce computing costs, and the RAT block mitigates the dependence on scene features and the influence of motion errors. These make RATIR-Net efficient and effective. Simulation experiments have been conducted to verify the proposed method.

Moving target indication (MTI) in bistatic synthetic aperture radar (BiSAR) is a promising task in both civilian and military fields. However, it suffers severe challenges under the influence of clutter. MTI in BiSAR mainly faces three challenges: 1) the range cell migration (RCM) of BiSAR is larger than that of monostatic synthetic aperture radar; 2) the clutter range-Doppler spectrum is severely extended; and 3) the clutter characteristic is closely related to geometrical configuration, with nonhomogeneity and nonstationarity. To solve these problems, based on single-channel BiSAR, a joint clutter suppression and MTI method is proposed. First, to ensure that the energy of an arbitrary target is concentrated in one range cell, RCM correction (RCMC) is completed by the first-order keystone transform (KT) and high-order RCMC. Then, an important step named increased dimension rotation (IDR) is applied, which mainly consists of two stages. One is increased dimension processing, which introduces a 2-D azimuth rotated time domain (ARTD). The other is rotation processing, which makes signals in one range cell rotated to 2-D ARTD. After that, the distribution characteristic between the moving target and stationary clutter in 2-D ARTD is analyzed. Next, an optimal filter for clutter suppression is designed according to the characteristics of signal distribution in 2-D ARTD. Furthermore, the corresponding inverse IDR processing and fractional Fourier transform (FrFT) are performed, and finally, MTI can be realized accurately. Generally, the proposed method has good robustness and low system complexity, and its effectiveness is proven by numerical simulations.

Despite the success of compressive sensing (CS) algorithms in sub-Nyquist SAR imaging, they lack a generalized ambiguity-depression capability and suffer from high computational complexity. To address these weaknesses, we propose a novel imaging approach that simultaneously suppresses ambiguity introduced by nonuniform sampling and non-ideal azimuth antenna patterns, while enhancing resolution performance for wide-swath sub-Nyquist SAR imaging. To improve efficiency, we employ the randomized block coordinate descent based on the accelerated fast iterative shrinkage threshold algorithm (FISTA) to obtain the desired ambiguity-free image. In the accelerated FISTA, we incorporate two measures to reduce computational complexity. Firstly, we analytically formulate the update stepsize using the singular-value perturbation theory, Kronecker product technique, and orthogonality of the uniformly discrete Fourier matrix. This enables efficient computation of the Lipschitz constant. Secondly, we expedite the computation of the gradient matrix through efficient frequency domain techniques. Numerical results using both simulated data and realistic SAR echo validate the effectiveness of our proposed method and demonstrate its advantages over existing algorithms.

Scale information is a significant index for target measurement by real aperture radar (RAR). However, the measured target scale information by RAR is inaccurate because of the limited angular resolution. In this paper, to enhance the scale measurement ability of RAR, a generalized hybrid regularization (GHR) approach is proposed by combining the generalized sparse (GS) regularization norm and the generalized total variation (GTV) regularization norm. On the one hand, the GHR approach is proposed to simultaneously enhance the angular resolution and the scale information of targets by combing the generalized regularization norms. The GS regularization norm can improve the reconstructed angular resolution due to its sparsity over the L1 norm. The GTV regularization norm can preserve the steep target contour because of its edge enhancement ability over the total variation (TV) norm. On the other hand, based on the GHR optimization function, an adaptive iterative reweighted (AIR) solver is proposed to reduce the number of manually selected regularization parameters, allowing for accurate scale information reconstruction. Simulations and experiments verify the performance of the proposed method. Based on the proposed approach and solver, the target scale information can be accurately observed.

Ship recognition in synthetic aperture radar (SAR) images is essential for many applications in maritime surveillance tasks. Recently, convolutional neural network (CNN)-based methods tend to be the mainstream in SAR recognition. Though considerable developments have been achieved, there are still several challenging issues toward superior ship recognition performance: 1) Ships have a large variance in size, making it difficult to recognize ships by using a single scale features of CNN. 2) The SAR ship’s large aspect ratio presents an obvious geometric characteristic. However, standard convolution is limited by the fixed convolution kernel, which is less effective in processing elongated SAR ships. 3) Existing CNN classifiers with softmax loss are less powerful to deal with intraclass diversity and interclass similarity in SAR ships. In this paper, we propose a task-specific hierarchically designed network with a spherical space classifier (HDSS-Net) to alleviate the above issues. Firstly, to realize SAR ship recognition with large size variation, a feature aggregation module (FAM) is designed for obtaining a feature pyramid that has strong representational power at all scales. Secondly, a FeatureBoost module (FBM) is devised to provide rectangular receptive fields to refine the features generated by FAM. Finally, a novel spherical space classifier (SSC) is proposed to expand the interclass margin and compress the intraclass feature distribution by fully taking advantage of the property of spherical space. The experimental results on two benchmark datasets (OpenSARShip and FUSAR-Ship) jointly show that the proposed HDSS-Net performs better than classic CNN methods and novel SAR ship recognition CNN methods.

Without sufficient data, the quantity of information available for supervised training is constrained, as obtaining sufficient synthetic aperture radar (SAR) training data in practice is frequently challenging. Therefore, current SAR automatic target recognition (ATR) algorithms perform poorly with limited training data availability, resulting in a critical need to increase SAR ATR performance. In this study, a new method to improve SAR ATR when training data are limited is proposed. First, an embedded feature augmenter is designed to enhance the extracted virtual features located far away from the class center. Based on the relative distribution of the features, the algorithm pulls the corresponding virtual features with different strengths toward the corresponding class center. The designed augmenter increases the amount of information available for supervised training and improves the separability of the extracted features. Second, a dynamic hierarchical-feature refiner is proposed to capture the discriminative local features of the samples. Through dynamically generated kernels, the proposed refiner integrates the discriminative local features of different dimensions into the global features, further enhancing the inner-class compactness and inter-class separability of the extracted features. The proposed method not only increases the amount of information available for supervised training but also extracts the discriminative features from the samples, resulting in superior ATR performance in problems with limited SAR training data. Experimental results on the moving and stationary target acquisition and recognition (MSTAR), OpenSARShip, and FUSAR-Ship benchmark datasets demonstrate the robustness and outstanding ATR performance of the proposed method in response to limited SAR training data.

Multistatic airborne synthetic aperture radar (MuA-SAR) systems can achieve high-resolution imaging in a short time by fusing observation data from multiple radar platforms. However, its imaging quality relies on a rigorous design of the spatial configuration (SC) of each platform, mainly including the relative spatial separation and velocity. The rigorously designed SCs make it difficult to obtain in actual flight and weaken the flexibility advantage brought by the airborne platforms. Therefore, it is meaningful and necessary to explore a new SC design method to obtain relaxed SCs under the condition of ensuring imaging quality. In this paper, to relax the limitations of SC, an optimal design method for MuA-SAR SC is proposed. First, the relationship between the spatial configuration, wavenumber spectrum (WS) distribution, and imaging performance is established, and it visually reveals the configuration limitations. Second, an optimized search space of SC is defined by the peak to sidelobe ratio (PSLR) to relax the space to compromised configurations. Finally, the SC design problem is transformed into a constrained multiple objective optimization problem (CMOP) which is solved by the multiple objective particle swarm optimization (MOPSO) algorithm. The simulation results show that the proposed method can still obtain the optimized SC beyond the strictly restricted configuration space, which expands the SC limitations of the MuA-SAR system.

Bistatic synthetic aperture radar (BiSAR) has been attracting worldwide attention because of its forward-looking imaging and high anti-interference. In harsh environment, it is vital for BiSAR to conduct extensive surveillance, imaging, and recognition of maritime ship targets. However, under the disturbance of sea waves, the ship target has an unknown and massive three-dimensional (3-D) rotation, so that its imaging projection plane (IPP) is also undetermined. Thus, high-dimensional random distortion appears in imaging results, making it difficult to recognize the target through two-dimensional distorted images effectively. To solve these problems, bistatic SAR maritime ship target 3-D image reconstruction method without distortion in local Cartesian coordinate (LCC) is proposed. In this paper, according to positions of scatterers and rotation parameters, significant differences of different scatterers of maritime ship targets have been found in bistatic range, Doppler centroid (DC), and Doppler frequency rate (DFR), which lays a solid foundation for the scatterer separation of ship targets. On this basis, a 3-D R-DC-DFR domain is constructed, and 2-D echoes of the maritime ship target are projected into R-DC-DFR domain to separate scatterers. Then, by remapping data of transmitter and receiver in R-DC-DFR domain to LCC, as well as evaluating their similarity metric, the optimal rotation parameters of the ship target can be obtained via the maximal similarity. Therefore, the image distortion caused by the unknown IPP has been removed, and 3-D image reconstruction of ship targets can be realized without distortion in the LCC. Furthermore, to evaluate performances of 3-D image reconstruction for different rotation parameters and bistatic configurations, 3-D reconstruction index is proposed and analyzed. Both point-targets and maritime ship targets are simulated to emphasize the effectiveness of the proposed method.

The sparse iterative covariance estimation (SPICE) algorithm is promising for hyperparameter-free sparse source location for time-division-multiplexing multiple-input multiple-output (TDM-MIMO) radar systems, with well-documented merits in resolution enhancement and sidelobe suppression. Regrettably, the method typically requires a large number of iterations to converge, each requiring high-dimensional matrix operations, rendering the existing batch SPICE method impractical and expensive to implement in hardware when dealing with massive TDM-MIMO observations. In order to enable real-time processing, this paper presents a sub-aperture-recursive (SAR) SPICE method, allowing for recursively refining the location parameters for each received (RX) block observation that becomes available sequentially in time. The proposed method not only offers the same benefits as the batch SPICE method, but also allows for a computationally efficient online processing, without the need for high-dimensional matrix operations, notably reducing the required hardware resources as well as processing time. We further present a high-throughput architecture for the resulting method on a XCZU15EG-FFVB1156 field-programmable gate array (FPGA). In combination with simulation results, we demonstrate the effectiveness through experimental data measured by a cascaded MIMO radar system with 12 transmit (Tx) and 16 Rx antennas, demonstrating that the computational time of resolving closely spaced sources on 256 predefined grid points can be processed in merely 12 ms.

Recently, the iterative adaptive approach (IAA) method has been adopted to allow for high-resolution direction of arrival (DOA) estimation of MIMO radar. In this paper, the computational complexity caused by traditional rough convergence criterion is reduced by following a selective-coordinate iterative strategy. First, we analyze the iterative termination criterion of the current IAA. Then, considering a novel criterion that defined by the absolute value of the adjacent iterative points of each coordinate, those coordinates that are not converged can be selectively iterated. In this way, unnecessary iterative calculations can be greatly reduced. Finally, the complexity of IAA and the proposed method are compared and analyzed in detail. Simulation and measured data illustrate that the proposed method offers a computational complexity reduction without loss of performance.

GEO spaceborne-airborne bistatic SAR (GEO SA-BiSAR) has flexible configuration and the ability of multi-looking imaging, so it has a good application prospect. The transmitting propagation delay is about 0.1s due to the 36500Km high altitude transmitter, so the motion of ground target and receiver in the propagation delay can't be ignored. This paper presents a range model under the “non-stop-and-go” assumption, which analyzes the transmitting and receiving process respectively, and we consider the movement of the target and receiving platform under long transmitting delay, and obtains a succinct and accurate range model. For the delay of each process, we establish the accurate numerical solution, the approximate solution and expanded of the time delay under the “non-stop-and-go” assumption separately. Simulation results show that our method can estimate the real propagation delay accurately for both stationary and moving targets.

Multistatic synthetic aperture radar (SAR) can realize 3D imaging of observation scenes by single navigation using distributed aperture. Meanwhile, due to the flexible baseline configuration of unmanned aerial vehicle (UAV), and has broad application prospects in remote sensing, surveying and mapping fields. However, the 3D reconstruction performance of multistatic SAR is closely related to its multistatic baseline. In this paper, a multistatic baseline design method is proposed to achieve optimal 3D reconstruction performance. Firstly, the quantitative relationship model between multistatic baseline and 3D imaging measurement matrix is established, and the cross-correlation value of measurement matrix is introduced as the evaluation index of reconstruction performance. Then, the multistatic baseline design problem is modeled as an optimization problem with an optimal crossrelation number. Finally, the differential evolution algorithm is used to obtain the multistatic baseline of the optimal design. Simulation results show that compared with the unoptimized multistatic baseline, the optimized multistatic baseline can obtain better reconstruction performance when the typical sparse recovery method is used for 3D imaging.

Receiving signals sequentially through multiple channels in azimuth, radar has the potential of forward looking high-resolution imaging. However, due to the limitation of platform size, its azimuth resolution is poor. In this paper, by considering the effect of platform motion and geometric distortion, a modified iterative adaptive algorithm(IAA) method is proposed to realize multichannel radar forward-looking superresolution imaging. Simulation results are illustrated to verify the effectiveness of the method.

Beam steering is widely applied in the airborne and space-borne SAR system to achieve a balance between large imaging area and high azimuth resolution. However, it leads to a linear variation of Doppler centroid in monostatic SAR, which is more complicated in bistatic SAR due to the contribution built by the transmitter and the receiver and results in Doppler spectrum aliasing. The spatial variance of the Doppler centroid of the beam steering bistatic SAR is analysed. And a modified preprocessing method for azimuth-variant bistatic SAR with beam steering mode is proposed to remove the Doppler spectrum aliasing by effectively increasing the sampling rate of azimuth time. The movement of the platforms is considered with cured trajectory.

Multistatic SAR based on unmanned aerial vehicles (UAV) swarm can simultaneously obtain the echo of the target from different perspectives. Thus it can synthesize a large aperture in a short time, realizing high resolution and high frame rate imaging of interested areas. But there is few imaging method for high frame rate UAV swarm SAR and the problem of over-lapping apertures affecting imaging quality remains to be re-solved. Therefore, this paper proposes a time-domain imaging method to solve the problems. Firstly, echo model of UAV swarm SAR is established and multistatic equivalent aperture distribution is analyzed. Then the echo is equalized to im-prove image quality. Next, coarse angular resolution images of each bistatic SAR pair in UAV swarm SAR is accurately formed. Finally, all bistatic images are recursively fused to get the final result. Simulation results verify the effectiveness and accuracy of the proposed method.

Airborne ultra-wideband synthetic aperture radar (UWB-SAR) is easily affected by motion error, leading to a decline in image quality. In this paper, a motion error compensation method based on coherent correlation function (CCF) is proposed, which can estimate the high-order components of motion error and reconstruct platform trajectory. Taking the peak sidelobe ratio (PSLR) of CCF as the evaluation function, the particle swarm optimization (PSO) algorithm is used to estimate the high-order components of the trajectory error. Then the trajectory is modified to compensate for the image blur caused by motion error. The effectiveness of the method is verified by numerical simulation.

Synthetic Aperture Radar (SAR) can perform all-time and all-weather observations and is wildly used in earth remote sensing. The sparsity-driven SAR imaging methods can reconstruct sparse scenes under down-sampling conditions, but they are unsuitable for non-sparse scenes. To reconstruct non-sparse scenes from under-sampled data and further improve the utilization efficiency of sampled data, this paper proposes a feature learning and SAR imaging method and implements it through a deep network. The imaging model is constructed firstly, where a feature-based sparsity regularization term is incorporated. Then, by unfolding the iterative solution derived via the Alternating Direction Multiplier Method (ADMM) algorithm, a CNN-based deep network is proposed to solve this imaging model. In the proposed network, convolution layers are used to represent and learn the scene feature prior knowledge. Simulation experiments verify the effectiveness of the proposed method.

Various imaging methods based on compressed sensing (CS) of synthetic aperture radar (SAR) have been proposed to reduce the sample size of echoes required for the imaging process. The unrolling technique further solves the inefficiency of conventional CS-based methods by mapping them into deep neural networks. However, most of these methods are based on sparsity prior of the scene or its transformation domain, which could be invalid for non-sparse scenes. To address this, we proposed a network utilizing the feature priors of the images instead of sparsity for non-sparse scene reconstruction of SAR, namely NSR-Net. We adopt learnable regularization terms in the CS model. Then the iterative solving process of the model is derived and unrolled into the proposed deep neural network to learn the best regularization terms from data. Simulation experiments verified the effectiveness of NSR-Net in the reconstruction of non-sparse scenes with down-sampled SAR echoes.

Synthetic aperture radar (SAR) can obtain two-dimensional images of the illuminated area, which is an important means for earth remote sensing and monitoring. However, due to the loss of azimuth data and system errors during the processing of data sampling, it is necessary to study the method for high-quality SAR image reconstruction from down-sampled data in the condition of measurement inaccuracy. Considering these factors, this paper proposes a sparsity-driven SAR imaging method based on general regularization and the sparse total least-squares (S-TLS) model and implements the method by an unfolded deep network. In the proposed method, general regularization can solve the problem of sparse sampling, and the S-TLS model is adopted to deal with measurement inaccuracy. Moreover, through the deep network implementation, the proposed is more time-efficient and can exploit more effective scene prior knowledge, making the proposed method suitable in practical applications. Experiments verify the effectiveness of the proposed method.

Synthetic Aperture Radar(SAR) image registration technology is widely used, but some problems are exhibited in the available technologies, such as feature extraction, loss of phase information, low computational complexity, and so on. To cope with these problems, in this paper, a method regards target Attributed Scatter Center (ASC) as the feature is proposed. ASC model highlights electromagnetic characteristics, which can better represent SAR images than optical characteristics. First, the parameter set of target ASC model is estimated by Iterative Shrinkage Thresholding Algorithm (ISTA), as a stable point feature registration descriptor of the image. Then, Nearest Neighbor Distance Ratio (NNDR) method is used to find the feature matching points of the reference image and the image to be registered. The matching accuracy is improved by the Random Sample Consensus (RANSAC) method. Finally, by transformation model and image offset, image registration and phase retention are realized. The simulation results demonstrate that the proposed method enables a better registration performance since the scattering characteristics are effectually utilized.

Although deep learning has achieved great success in automatic target recognition, the model needs a large number of labeled samples for training. In real life, it is a time-consuming and laborious work to label unlabeled samples, so how to find unknown classes from a large number of unlabeled samples has aroused widespread concern. In this paper, we study how to discover unseen class from an unlabeled image set under the assumption that there are samples related to the unseen class but of different classes as prior knowledge. The Automatic Unseen Class Discovery (AUCD) algorithm is proposed in this paper, which mainly solves the problem of unseen class discovery from two aspects, one is how to actively form clusters according to their classes for unknown class samples, and the other is how to obtain the number of formed clusters. Several experiments based on the Moving and Stationary Target Acquisition and Recognition (MSTAR) dataset prove the effectiveness of the proposed approach in the field of unseen class discovery.

Localization plays an important part in the application field of bistatic SAR (BiSAR). The accuracy of traditional localization method for BiSAR depends on the measurement precision of platform. The lower measurement precision, the higher localization error of targets. However, due to the size, cost, and technical limitations of existing attitude measurement devices, it is difficult to satisfy the requirement of high-precision target positioning in BiSAR. This paper proposes a blind locali-zation method based on monostatic equivalent model for BiSAR, which has high tolerance and low sensitivity for measurement error. First, the BiSAR is equivalent to monostatic SAR system based on the principle of Equivalent Phase Center (EPC). Next, the Range-Doppler localization model is applied to fix position for EPC and the relative position of multi-targets are obtained subsequently. At last, establishing localization equation with bistatic range of multi-targets, and the Newton iteration algorithm is devoted to solving localization model constructed above. The effectiveness of the localization method proposed in this paper has been demonstrated and illustrated by numerical simulations.

Remote sensing images (RSIs) play an important role in a wide range of applications. However, they are frequently contaminated by multiple kinds of noises and existing methods are mostly applied to suppressing single noise type and performs poorly for various noises. To deal with above deficiencies, we propose a learning-based multi-type noise suppressing method (MNSM). Firstly, “Parallel” denoising approach is utilized to obtain partially denoised images that supply sufficient information for the subsequent fusion task. Mean-while, the noise recognition net identifies noise type and adjusts the brightness of every partially denoised image, realizing the adaptivity for different noises. The fusion net lastly merges these images to acquire one clean image. Experimental results show that this approach obtains higher peak-signal-to-noise ratio (PSNR) than existing methods.

Sparse linear array synthetic aperture radar (SAR) is widely used in near-field 3-D imaging, which can reduce the volume of data and the cost of acquisition. However, the sparsity of array elements means the missing rows or columns of data, leading to the high-level sidelobes and the aliasing of targets. Therefore, in this paper, we develop a near-field 3-D imaging method for non-uniform sparse linear array SAR based on matrix completion (MC), where a rotated expansion algorithm is proposed to make the non-uniform sparse data af-ter range compression satisfies the requirements of the MC. The MC technique is applied to the signal slice of each range cell to reconstruct the echo signal, and the back-projection algorithm (BP) is used to achieve 3-D high-resolution imaging of targets. Moreover, simulations and near-field experi-ments validate the feasibility and superiority of the proposed method.

Multistatic Synthetic Aperture Radar (Mu-SAR) can obtain rich target information through multi-platform collaboration, and topology configuration is one of the most important factors that affecting the imaging performance. In this paper, a Mu-SAR topology design method is proposed. First, the echo of Mu-SAR is analyzed in wavenumber domain, the relationship between wavenumber spectrum and topology configuration is deduced. Then, a topology design method based on wavenumber spectrum range extension is proposed to obtain topology configuration that can achieve high resolution imaging in range direction. Finally, through numerical simulation, the effectiveness of the proposed method is verified.

In order to improve the angular resolution of scanning radar, a lot of super-resolution algorithms have been developed in recent years. However, the super-resolution performance is affected by many factors due to the ill-conditioned nature of inverse problem. In this paper, from the perspective of numerical simulation, we illustrated the influence of the antenna pattern sidelobes on the condition number of the convolution matrix, and then analyzed the influence on the performance of the super-resolution algorithm.

Clutter suppression is the premise of moving target detection and imaging. We propose a cancellation method for GEO Spaceborne–Airborne Bistatic SAR (GEO SA-BiSAR). Firstly, according to the range history under the “non-stop-and-go” assumption, we establish a multi-channel echo signal model. Then, according to the range from the high orbit transmitting station to the target, we compensate the phase of the echo signal. Next, by analyzing the multi-channel phase relationship, we propose an improved Displaced Phase Center Antenna (DPCA) method suitable for GEO SA-BiSAR, and then we analyze the result and performance of DPCA. Finally, the effectiveness of this clutter suppression method is verified by numerical simulation.

In the field of remote sensing, harbor detection in SAR images has an important application prospect. However, the complex coastline of SAR images increases the difficulty of harbor detection. In response to this problem, a cascaded harbor detection (CHD) method for SAR image based on corner and coastline features is proposed in this paper. First, coast-line is extracted from SAR image by sea-land segmentation. Then, in the first step rough detection, corner detection is performed on the coastline and the detected corners are automatically clustered to locate the harbor candidate areas. Finally, the second step precise detection is carried out on the coast-line of harbor candidate areas, where coastline feature detection is completed by using corners again to remove the fake harbor targets in harbor candidate areas. Experimental results based on satellite-borne SAR data prove the proposed CHD method enjoys a preferable detection performance compared with existing harbor detection methods.

To address scanning radar forward-looking sea imaging, this paper proposed a super-resolution method based on Weibull distribution. The proposed method in our work introduced the generalized Gaussian distribution and Weibull distribution to represent the prior distribution of the target and the sea clutter respectively, which are more suitable for actual sea imaging. And the corresponding objective function was derived under the MAP framework. In order to overcome the objective function's nonlinearity, this paper adopt Newton-Raphson iterative method to resolve it. Finally, through simulations, which indicates that the proposed method has superior imaging performance compared with other traditional methods for sea imaging.

High sidelobe of strong scattering points usually submerges weak targets nearby and affects the quality of SAR image. Therefore, SAR image usually requires sidelobe control. Common window functions have limited improvement on PSLR performance when the image resolution is required to be guaranteed. Combining Min-Max weighted ISL technique, this paper proposes an azimuth low sidelobe window function design method for SAR. Simulation results show that PSLR of the designed window is nearly −10dB lower than hanning window with a −45dB ISL level, and main lobe width is almost equal to hanning window.

Exploiting the sparse property of the target of interest to achieve super-resolution imaging has been applied to real aperture radar (RAR) forward-looking imaging field. In this paper, we proposed a regularized FOCUSS method to realize RAR forward-looking super-resolution imaging. In addition, we discussed the influence of initialization on the imaging result, and selected the most suitable initialization for RAR forward-looking super-resolution imaging. Compared with the traditional sparse method based on Majorize-Minimization, our proposed algorithm has faster convergence speed under the same parameters condition.

The real aperture radar (RAR) system can present full-view observation capability, but the coarse azimuth resolution restricts its application. Therefore, various super-resolution deconvolution methods are widely used in the real aperture super-resolution imaging field. But conventional deconvolution approaches only rely on amplitude information of antenna pattern profile, it will behave worse when forward-looking imaging with high speed or squint imaging. This paper analyzes the influence of phase and constructs a corresponding amplitude-phase model to resolve this problem. Finally, the effectiveness of the proposed amplitude-phase convolution model for forward-looking imaging with high speed or squint imaging is verified by simulations.

With the continuous development of imaging radar technology, the bandwidth of transmission signals continues to increase. The geometrical theory diffraction (GTD) model can describe target scattering characteristics under wideband electromagnetic wave irradiation, which is significant for target detection and recognition. Existing target scattering center parameter estimation methods based on compressed sensing (CS) theory require sparse imaging scenes, which is difficult to meet in actual radar data. To solve this problem, this paper proposes a method for estimating high-value scattering centers parameters based on energy trajectory extraction. The estimation result of this method is accurate with fewer operations than CS-based method, and this method reduces the requirement for target sparsity. And use numerical simulation to prove the effectiveness of the method.

SAR and optical images registration is a key step for remote sensing image processing, match navigation and information fusion. Although there are many methods for SAR images registration, their performance will decrease between SAR and optical images. Moreover, these algorithms suffer from lack of matching pairs of the feature points and uneven distribution between SAR and optical images. Therefore, they cannot accurately achieve the registration between optical and SAR images. To solve the above deficiencies, we propose an efficient image registration approach based on SAR and optical image-scale invariant feature transform (SOI-SIFT). Firstly, a linear edge enhancement based on gray feature and histogram equalization is introduced. In this stage, we enhance the edge features of the image so that the number of image feature points can be greatly increased. Then, for feature points purification, we use fast sample consensus algorithm to filter duplicate and wrong matching feature points. SOI-SIFT can be more adapted to the heterogeneous image matching. Experimental results have shown the superiorities of the proposed method.

Synthetic Aperture Radar (SAR), an all-weather microwave remote sensing technology, is widely used in environmental monitoring, earth resource surveys and other fields, especially in the military field to achieve monitoring purposes. In the monitoring process, it is crucial to achieve real-time imaging of the target scene. Therefore, in this paper, an efficient SAR real-time imaging system is designed to achieve timely parallel processing of the scene when the radar platform scans the target scene, so as to reduce the consecutive frame cycles of real-time imaging. In which, the signal processing board structure consists of two FPGAs and six DSPs is used to quickly implement high precision SAR real-time imaging for airborne radar, so that the radar system can display the target image in time with the platform moving and check whether the current echo is valid. In addition, the architecture proposed in this paper can be extended in case the continuous frame period for real-time imaging is short.

Short-time pulse train signal has narrow pulse width and high peak power, which enables it with great potential for airborne radar target detection. Clutter interference always exists in complex detection environment, thus the design of the anticlutter emission waveform decides the accuracy of target detection. In this paper, a secondary optimization method based on maximum signal-clutter-to-noise ratio (SCNR) criterion and minimum squared error criterion is proposed for short-time pulse train signal form. Firstly, a maximum SCNR model solved by the Lagrange multiplier method is constructed based on prior information. Secondly, according to the short-time train signal form, the time-domain form of the optimal transmission signal is obtained by minimizing the weighted square error of the optimal spectrum and the train spectrum and the integrated side lobe of the train signal. Finally, simulation results show that the designed waveform is capable of anti-clutter interference and can effectively improve the SCNR in the clutter environment.

Recently, super-resolution techniques have been widely used in real aperture radar superresolution imaging. In this paper, we propose a fast sparse superresolution algorithm which is based on majorize-minimization(MM) method to realize fast superresolution imaging of sparse targets in radar forward-looking area. First, we establish a model of rader forward-looking imaging and analyze the echo signal. Second, we use the majorize-minimization (MM) method to obtain the real target distribution. Due to the expensive computational cost of MM algorithm, we proposed an fast matrix inversion approach which is based on divide and conquer strategy. The superior performance of the proposed method is verified by simulations.

Multi-view SAR images contain richer target information than single-view, which is beneficial to synthetic aperture radar automatic target recognition (SAR ATR). It is a huge challenge to select the best observation viewpoints and the most suitable flight path for multi-view SAR ATR in an unknown environment. Therefore, we propose a multi-view SAR ATR optimal observation path planning method in this paper. The geometrical and the optimization mathematical models based on the task requirements are constructed, and the convolutional neural networks with two inputs are designed as the base classifier. An autonomous path planning method forms the best observation path planning in the absence of global information of the surroundings. Thus the selection of the optimal viewpoint for multi-view SAR ATR is solved by the path search algorithm. The multi-view SAR images are collected on the solved optimal viewpoints, and the final recognition result is obtained by the base classifiers ensemble. Experimental results based on the moving and stationary target acquisition and recognition (MSTAR) dataset have shown that the proposed method obtains superiority in optimal observation path planning.

In synthetic aperture radar (SAR) images, there are a large number of dense multi-scale objects, especially dense multi-scale ships docked along the coast. Existing object detection methods are difficult to simultaneously detect dense multi-scale objects in complex background. A novel method for dense multi-scale object detection in SAR images based on Feature-Transferable Pyramid Network (FTPN) is proposed in this paper. In the stage of feature extraction, the feature maps of each layer are connected effectively and the feature maps of various scales are extracted. This method can extract the features of dense multi-scale objects more effectively, so as to realize simultaneous detection of dense multi-scale objects in SAR images. Experiments on SSDD dataset, AIR-SARShip-2.0 dataset and Gaofen-3 dataset show that the proposed method can achieve dense multi-scale object detection, and the overall performance is better than the state-of-the-art methods.

Stochastic radiation radar (SRR) transmits space-time two-dimensional stochastic signals to achieve superresolution imaging and can overcome the geometric acquisition limitations of traditional synthetic aperture radar (SAR). However, the resolution of an SRR system is limited by the number of effective singular values of its stochastic radiation field (SRF). In this letter, an interpulse frequency hopping accumulation (IFHA) method is proposed to improve the resolution of an SRR system. First, an SRR signal model is introduced. The rank of the traditional SRF generation method is quantitatively analyzed and is limited by the number of transmitting array elements. Second, an IFHA method is proposed to increase the number of the effective singular values of the SRF matrix, which can improve the superresolution imaging performance of SRR. Finally, the simulation results verify the effectiveness of the proposed method.

Benefiting from the capability of forward-looking imaging, ability of receiver radio silence and resistance of jamming, bistatic SAR has extensive potential applications. However, due to its independent dual platform movement, motion error of bistatic SAR is usually more complicated than monostatic SAR and could easily exceed range resolution cell, which will result in both azimuth and range defocusing if not properly compensated. In this paper, an autofocus scheme for bistatic SAR considering cross-cell residual range migration is proposed. In this scheme, cross-cell residual range migration is firstly compensated by estimating bistatic range error through time-frequency analysis. Meanwhile, azimuth phase error is coarsely compensated by using bistatic range error estimation result. Secondly, maximum image sharpness autofocus method is cascaded to further compensate for azimuth residual phase error and enhance the quality of bistatic SAR image. The effectiveness of proposed scheme is verified by simulation and experiment results.

For real aperture radar, its azimuth resolution is much coarser than the range resolution after pulse compression, super-resolution algorithms are desired to enhance its azimuth resolution. However, the super-resolution algorithms must require enough azimuth sampling to ensure its performance. When wide scanning scope or dense azimuth sampling, the amount of data will increase significantly, which brings large computational burden to super-resolution processing. To cover this problem, we propose a two-step dimension reduction strategy. Firstly, by using linear sketching technology, the high-dimensional matrices are projected to the low-dimensional space, thus accelerating the matrix-matrix multiplications in super-resolution algorithms. Secondly, exploiting Sherman-Morrison formula, we further realized the acceleration of the matrix inversion in super-resolution algorithms. The proposed two-step acceleration strategy in our work is applicable to the existing deconvolution super-resolution algorithms, including regularization methods, Bayesian methods. It can be verified by simulation and experimental data that the proposed accelerated algorithms have advantages in computing time without losing the quality of super-resolution imaging.

Convolutional neural networks (CNNs) have dominated the synthetic aperture radar (SAR) automatic target recognition (ATR) for years. However, under the limited SAR images, the width and depth of the CNN-based models are limited, and the widening of the received field for global features in images is hindered, which finally leads to the low performance of recognition. To address these challenges, we propose a Convolutional Transformer (ConvT) for SAR ATR few-shot learning (FSL). The proposed method focuses on constructing a hierarchical feature representation and capturing global dependencies of local features in each layer, named global in local. A novel hybrid loss is proposed to interpret the few SAR images in the forms of recognition labels and contrastive image pairs, construct abundant anchor-positive and anchor-negative image pairs in one batch and provide sufficient loss for the optimization of the ConvT to overcome the few sample effect. An auto augmentation is proposed to enhance and enrich the diversity and amount of the few training samples to explore the hidden feature in a few SAR images and avoid the over-fitting in SAR ATR FSL. Experiments conducted on the Moving and Stationary Target Acquisition and Recognition dataset (MSTAR) have shown the effectiveness of our proposed ConvT for SAR ATR FSL. Different from existing SAR ATR FSL methods employing additional training datasets, our method achieved pioneering performance without other SAR target images in training.

Ship target segmentation plays an important role in synthetic aperture radar (SAR) image interpretation. However, existing segmentation methods for marine SAR images have the problem of inaccurate edge segmentation, a concern for real-world applications. In this letter, we propose a clustering center shifted adaptive target segmentation (CCSATS) method. Firstly, the proposed clustering center shift method is used to update the clustering centers of each iteration, which can quickly and accurately capture ship pixels. Then, based on regional homogeneity coefficients, we define a new similarity measurement criterion with two adaptive weight factors to ensure the homogeneity of segmentation results. Finally, neighborhood patches are used to represent pixel information, which can reduce the influence of speckle noise and enhance the target edge fitting ability. Our segmentation results of measured SAR images show that the proposed method effectively ensures segmentation accuracy. Compared with other existing methods, the proposed target segmentation method achieves better edge capture performance.

The generalized sparse iterative covariance-based estimation ($q$-SPICE) algorithm was recently introduced for scanning radar applications, resulting in substantial improvements in the angular resolution and quality of the processed images. Regrettably, the computational complexity and storage cost are high and quickly increase with growing data size, limiting the applicability of the estimator. In this letter, we strive to alleviate this problem, deriving a beam-updating$q$-SPICE algorithm, allowing for efficiently updating of the sparse reconstruction result for each online radar measurement along the scanned beam. The resulting method is a regularized extension of the current online$q$-SPICE implementation, which not only offers constant computational and storage cost, independent of the data size, but also provides enhanced robustness over the current online$q$-SPICE. Our experimental assessment, conducted using both simulated and real data, demonstrates the advantage of the beam-updating$q$-SPICE method in the task of sparse reconstruction for scanning radar.

Synthetic aperture radar (SAR) imaging with sub-Nyquist sampled echo is a challenging task. Compressed sensing (CS) has been widely applied in this case to reconstruct the unambiguous image. The CS-based methods need to set the iterative parameters manually, but the appropriate parameters are usually difficult to obtain. Besides, such methods require a large number of iterations to obtain satisfactory results, which seriously restricts their practical applications. Moreover, the observation scene of SAR is not sparse in some cases. In this paper, we aim at proposing an efficient and effective imaging method for non-sparse observation scenes with reduced data. Firstly, considering the characteristics of non-sparse observation scenes in SAR imaging, we model the SAR imaging problem as a joint low-rank and sparse matrices recovery problem. After that, the iterative alternating direction method of multipliers (ADMM) to solve the above problem is unrolled into a layer-fixed deep neural network with trainable parameters, in which the learnable parameters are layer-varied. The threshold parameters, as well as the weight parameter between the sparse part and low-rank part of each layer, are learned adaptively instead of manually tuned. Experiments prove that the proposed LRSR-ADMM-Net is capable of reconstructing the non-sparse observed scene with high efficiency and precision. Particularly, the proposed LRSR-ADMM-Net yields better reconstruction performance while maintaining high computational efficiency compared with the state-of-the-art iterative recovery methods and the trainable sparse-based network methods.

Video synthetic aperture radar (SAR) receives more and more attention in recent years because it can provide continuous images of the observed scene. However, the enormous data of video SAR to obtain the multiframe images bring big challenges to its transmission, storage, and processing, especially for small unmanned aerial vehicle (UAV) platform. In this article, we aim at proposing an efficient video formation method for video SAR systems with reduced data. First, the characteristics of video SAR observed scene are analyzed. It is found that the observed scene with multiple frames can be modeled as the sum of a low-rank tensor and a sparse tensor efficiently. After that, the video formation problem for video SAR is modeled as a joint low-rank and sparse tensors recovery problem. Finally, an efficient tensor alternating direction method of multiplier is proposed to obtain the final SAR video. Compared with the traditional frequency- or time-domain imaging methods, the amount of data samples can be greatly reduced. On the other hand, the proposed method outperforms the state-of-the-art SAR imaging methods with reduced samples, including the joint low-rank and sparse matrices recovery method and the low-rank tensor recovery method. Numerical simulations validate the effectiveness of the proposed method.

Geosynchronous spaceborne–airborne bistatic synthetic aperture radar (GEO-SA-BiSAR) consists of a GEO transmitter and airborne receiver, which has extensive application prospects in both civilian and military fields for its ability to generate high-resolution images of the ground target with frequent coverage and abundant scattering information. However, the Doppler bandwidth in this configuration exceeds the transmitted pulse repetition frequency (PRF), which leads to sub-Nyquist sampling. To solve this problem, a multireceiving technique has been applied to the receiver to increase the equivalent sampling rate and reconstruct an unambiguous image. In this article, we take a different approach to recover the unambiguous image for GEO-SA-BiSAR with fewer receiving channels. First, the accurate echo model is established based on the “non-stop-and-go” propagation delay model to lay the foundation of accurate imaging. After that, the GEO-SA-BiSAR imaging problem is modeled as a problem of joint sparse and low-rank matrices’ recovery. To reduce the computing time of the traditional recovery method, a modified alternating direction method of multipliers (M-ADMM) is proposed, where the computation and storage of the computational expensive observation matrix are avoided. Furthermore, an M-ADMM method with multiple receiving channels, which combines the recovery theory and multireceiving information, is also proposed to handle the severe sub-Nyquist sampling echo of GEO-SA-BiSAR. Simulation results reveal that the proposed method can recover the original image scene with high computational efficiency. Meanwhile, the number of receiving channels can be reduced compared with the multireceiving technique.

With the maturity of synthetic aperture radar (SAR) technology, the problem of imbalanced data has gradually emerged. This problem makes it difficult for the automatic target recognition (ATR) model to properly learn the classification boundaries of majority and minority category target samples. In this article, we propose an ATR model with new architecture, called the cost-sensitive awareness-based automatic target recognition (CA-ATR) model, which provides an effective way of solving the problem of imbalanced data. Aimed at the two issues caused by imbalanced data on ATR models, the proposed method solves the problems from both the data and algorithm levels. At the data level, CA-ATR avoids adverse correlations among the target samples through different oversampling methods. By making the ATR model cost-sensitive, the proposed method also avoids the empirical risk preference of the ATR model for majority category target samples at the algorithm-level. At the same time, CA-ATR can autonomously learn different cost-sensitive awareness from different imbalanced data sets. The awareness enables the ATR model to more accurately learn the classification boundaries between target samples that belong in different categories. Several experimental results show the superiority of the proposed approach based on the Moving and Stationary Target Acquisition and Recognition (MSTAR) data set. Compared with other imbalanced learning methods, the proposed method is able to solve different types of imbalanced data problems.

Since there are differences in the natural frequency of various synthetic aperture radar (SAR) target samples in reality, the problem of imbalanced data on the automatic target recognition (ATR) model has gradually appeared in recent years. The problem makes the classification boundary learned by the ATR model often fuzzy or even wrong. In this article, an SAR target sample generation method was proposed, called demand-driven generative adversarial nets (DDGANs), which provided an effective way to implement imbalanced data learning. When the imbalanced data exacerbated the deterioration of the minority category target samples distribution, the proposed method generated samples to alleviate this negative impact. The proposed method innovatively used two convolutional neural networks to form the discriminator of DDGAN. Among them, a convolutional neural network was used to determine whether the generated sample is real or fake. Moreover, another convolutional neural network can simultaneously dig out the generation demands of different categories of target samples when recognizing the generated samples. The generation demands enabled DDGAN to allocate different generation capabilities to different target samples on demand, thereby alleviating the negative impact of data imbalance. At the same time, DDGAN can autonomously learn the generation demands from imbalanced training sets. Several experimental results based on the Moving and Stationary Target Acquisition and Recognition (MSTAR) dataset showed the advantages of DDGAN. Compared with existing imbalanced learning algorithms, the proposed method had obvious superiority in recognition performance and data generation efficiency.

Microwave photonic synthetic aperture radar (MWP-SAR) offers a larger signal bandwidth than conventional SAR, and its theoretical resolution can be improved to centimeter level. To achieve same order of magnitude resolution in azimuth direction, long synthetic aperture is always required, which results in extremely high requirements for the accuracy of imaging procedure. Compared with conventional SAR imaging algorithms, two major problems should be considered: 1) The scattering characteristics of target might vary from the frequency of signal and the angle of incidence, which seriously affects the coherence of received echo. 2) The imaging of MWP-SAR system is more sensitive to motion errors and therefore requires higher accuracy of motion compensation processing. To solve the above issues, a high-precision imaging method for MWP-SAR is proposed. First, based on the attribute scattering center(ASC) model, this paper analyzes the target-scattering characteristics with different frequencies and incident angles. Then, according to the influence of scattering phase on MWP-SAR imaging, an optimal sub-aperture division algorithm is proposed to guarantee the coherence of each sub-aperture data and better imaging results. Furthermore, to compensate for the effects of high-order motion errors, this paper proposes a motion error estimation algorithm based on sub-image registration, where the full aperture high-order motion errors can be divided into multiple linear components, and the flight trajectory of platform can be accurately reconstructed. Finally, a full-aperture time-domain high-precision imaging method is presented, and the effectiveness of the proposed formation is verified by both simulation and actual airborne MWP-SAR data processing.

The global navigation satellite system (GNSS)-based passive radar shows potential in permanent maritime surveillance. In this paper, the GNSS signals are exploited for vessel target imaging. From the obtained radar image, meaningful information about the vessel, such as its shape, position, length, and orientation can be extracted. In addition, the vessel is observed from different angles by spatially diverse GNSS satellites, and the multistatic geometry enables to enhance the imagery quality. The main drawback of GNSS-based passive radar stays in its limited power budget. And the inaccessible motion makes the noncooperative vessel smeared using conventional radar imaging methods. To address the problems, at first, each bistatic echo over a long observation time is integrated in range and Doppler (RD) domain after removing the two-dimensional migrations. The signal-to-noise ratio can be increased after the step. Then, with respect to a particular target velocity, the local Cartesian plane is constructed, and the multiple RD maps are projected and combined in the plane to obtain the multistatic image. In view of the inaccessibility of target kinematic parameters, such imaging processing is modeled as an optimization problem, where vessel’s velocity is set as decision variable and the aim is to minimize the image entropy. Finally, particle swarm optimization (PSO) algorithm is applied to solve the optimization problem, after which a well-focused vessel image can be obtained. In May 2021, we have successfully carried out the world’s first BeiDou-based passive radar maritime experiment, and effectiveness of the proposed method is verified against the experimental data.

Although deep learning methods have achieved great success in synthetic aperture radar automatic target recognition (SAR ATR), their accuracies decline sharply as new classes are learned, which is known as catastrophic forgetting. The overlapping or confusion between the representations of new and old classes in the feature space is the main cause of catastrophic forgetting. In this paper, the Incremental Class Anchor Clustering (ICAC) is proposed to address this issue. ICAC solves this problem from three perspectives: first, how to learn the new classes; second, how to enable the model to recognize and classify the old classes; third, how to solve the imbalance between old classes and new classes. To learn the new classes, ICAC adaptively adds new anchored class centers for new classes, and the features of each new class will be clustered around the corresponding anchored class center. To enable the model to recognize and classify the old classes, ICAC stores some exemplars for the old classes to ensure the classification ability of the old classes without losing the old class centers in the feature space. At the same time, ICAC adopts knowledge distillation to further alleviate catastrophic forgetting. To solve the imbalance between old classes and new classes, ICAC proposes a learning strategy named Separable Learning (SL), which computes the losses of the old and new exemplars separately and then adds the two losses to make a gradient descent. Experiments on the MSTAR dataset and OpenSARShip dataset demonstrate the effectiveness of this method in SAR automatic targets recognition.

This article puts forward a BeiDou-based passive multistatic radar (PMR) maritime moving target (MMT) detection technique via space–time hybrid integration (STHI) processing. Compared with passive bistatic radar (PBR), the utilization of multiple satellites provides an improvement in MMT detection performance, together with the capabilities of localization and velocity estimation. However, the multiple satellite transmitters cause the differences principally in bistatic range and Doppler centroid (DC) of the MMT. To integrate the PMR echoes, the biggest challenge is the two differences that need to be handled. In the proposed technique, first, the centroid-compensated keystone transform (CCKT) is proposed and applied to each PBR echo. It not only corrects range cell migration (RCM) but also equalizes the DC to the same. Then, the long-time integration is performed on each PBR echo, after which it is integrated into the range-Doppler frequency rate (DFR) domain. Finally, in order to settle the difference in bistatic range, an MMT position and velocity domain (i.e., the$X$–$Y$–$V$domain) is constituted. The obtained multiple range-DFR maps are projected to the$X$–$Y$–$V$domain, and then, the effective integration of multistatic echoes can be implemented. The final STHI result allows detecting the MMT reliably. Meanwhile, according to the 3-D position where the MMT is located in the$X$–$Y$–$V$domain, the MMT can be localized, and its velocity can be estimated simultaneously. In May 2021, we have successfully carried out the world’s first BeiDou-based PMR MMT detection experiment, and the experimental results are given to prove the effectiveness of this technique.

Synthetic aperture radar (SAR) is an important means for target surveillance through reconstructing the microwave image of the observation area. However, under the condition of low signal-to-clutter ratio (SCR), such as a strong sea clutter situation, it is difficult to surveil targets from SAR images acquired by the traditional matched filter-based imaging methods. To improve the target surveillance performance of SAR, this article proposes a target-oriented SAR imaging method, which can enhance the desired target and improve the SCR in the reconstructed SAR images. By separating the target area from the clutter area, we first establish a target-oriented SAR imaging model, where the generalized regularization is used to characterize the features of the target, contributing to the improvement of SCR in the reconstructed image. Then, the imaging model is solved through a deep network, MF-ADMM-Net, which is obtained by unfolding an alternating direction method of multipliers (ADMM)-based iterative solution. In addition, the training strategy is formulated with the consideration of complex values. Experiments are conducted to verify the performance of image reconstruction and SCR improvement of the proposed method, and comparisons show the superiority of MF-ADMM-Net in effect and efficiency.

Synthetic aperture radar (SAR) can provide high-resolution electromagnetic backscattering images of the illuminated area, playing a significant role in various applications. However, achieving focused SAR images is challenging under sparse sampling and phase error conditions. By exploiting the sparsity or compressibility priors, the state-of-the-art sparsity-driven SAR imaging methods can reconstruct images under the condition of sparse sampling. However, the handcrafted priors used in these methods limit the imaging performance, and the iterative solution schemes reduce the computational efficiency. Besides, the measurement inaccuracy introduced by the phase error also degrades the reconstruction performance of the sparsity-driven imaging methods. To address these issues, a deep network for SAR autofocus imaging is proposed, which alternately performs image reconstruction and phase error estimation. When performing image reconstruction, the sparsity-cognizant total least-square (S-TLS) model is introduced to handle the problem of measurement inaccuracy, contributing to robust reconstruction performance under the condition of phase error. During the implementation of the deep network, a feature transform operator is used to realize data-driven prior knowledge learning and overcome the limitations of handcrafted priors. Moreover, the deep network approach can significantly improve computational efficiency. Experiments on simulated and real data verify the effectiveness and efficiency of the proposed method.

Clutter suppression is a challenging task in synthetic aperture radar-ground moving target indication (SAR-GMTI). In general, sufficient secondary samples are not easily acquired due to the nonstationary and nonhomogeneous characteristics of bistatic SAR (BiSAR) clutter, resulting in worse clutter suppression results. Recently, space–time adaptive processing based on sparse recovery (SR-STAP) has been developed since its better clutter suppression performance with less samples. However, since the off-grid problem in space–time domain caused by BiSAR’s separate configuration, existing SR-STAP would suffer from severe performance degradation. To address this problem, a clutter-ridge matched STAP (CRM-STAP) method for BiSAR nonstationary clutter suppression is proposed. First, clutter distribution modeling with arbitrary BiSAR configuration is applied to accurately obtain the clutter ridge in space–time domain. Then, keystone transform and time-division processing are applied to correct range cell migration and eliminate Doppler frequency migration, respectively. Next, to solve the off-grid problem, the CRM dictionary is reconstructed via adaptive gradient method, which is established along the direction of clutter ridge and its orthogonal direction. Then, with the constructed CRM dictionary, the clutter covariance matrix (CCM) estimation process is transformed to a multimeasured vector optimization problem, and it can be directly solved by the sparse Bayesian learning algorithm. Finally, based on the estimated CCM, the CRM-STAP filter is built to suppress the nonstationary clutter effectively. Compared with the existing STAP and SR-STAP methods, this method can avoid the performance degradation in clutter suppression caused by the off-grid problem and overcomes the strong nonstationary problem of BiSAR clutter in heterogeneous environments. In October 2020, we have successfully carried out the world’s first airborne BiSAR-GMTI experiment, and the experimental results are given to verify the effectiveness of this method.

Bistatic forward-looking synthetic aperture radar (BFSAR) is a kind of bistatic SAR system that can image forward-looking terrain in the flight direction of the receiver. Current literature and reports about BFSAR mainly concentrate on the stationary scene and ground-moving target imaging. Unlike stationary and ground-moving targets, the translational and rotational movements of ship targets usually lead to complicated range cell migration (RCM) and Doppler frequency migration (DFM). Moreover, the characteristics of RCM and DFM for different scattering points of the ship target are significantly different, i.e., the characteristics of the RCM and DFM are 2-D spatial variation, ultimately leading to severe defocusing of ship target in the SAR image. To solve these problems, a kind of hybrid SAR-ISAR imaging formation is proposed for BFSAR ship target imaging. First, to solve the problem of the Doppler ambiguity caused by the forward-looking mode of the receiver, an efficient ambiguity estimation method based on the minimum entropy criterion is presented. Then, keystone transform and range alignment processing can be applied to correct the spatial variant range walk and higher order RCM, respectively. Moreover, in order to obtain a high-resolution and well-focused image after translational compensation, a new method based on the fractional Fourier transform (FrFT) and the Wigner–Ville distribution (WVD) is proposed, where FrFT is applied to separate the multiple main scattering points in each range cell, and WVD is applied to obtain the high-resolution time–frequency distribution of each scattering point. Compared with the conventional ISAR range-Doppler (RD) algorithm and time–frequency estimation-based imaging methods, this method not only has no cross terms but also has high processing accuracy and better antinoise performance.

Clutter suppression in synthetic aperture radar (SAR) is one of the urgent and attractive problems in ground moving target indication (GMTI) application. With separate transmitter and receiver, ground clutter is nonstationary in bistatic forward-looking SAR (BFSAR), which directly leads to the inaccurate clutter covariance matrix (CCM) estimation. As a consequence, the traditional space-time adaptive processing (STAP) will suffer from a serious performance deterioration. In this article, an optimally matched space-time filtering (MSTF) technique is proposed to suppress nonstationary clutter for BFSAR systems. The main idea of the proposed method is to directly design and generate a suppression filter in space-time domain, whose space-time frequency response is matched with clutter spectrum. To construct the matched space-time filter, clutter modeling with arbitrary BFSAR configuration is first proposed to acquire space-time information of clutter spectrum. And then, the suppression filter can be designed and the design process is transferred into a constrained optimization problem (COP), according to the obtained clutter space-time information. Finally, the particle swarm optimization (PSO) algorithm is applied to solve the COP and obtain the optimal solution, i.e., the desired matched space-time filter weight, for BFSAR nonstationary clutter suppression. Since the generation of the designed filter circumvents CCM estimation, the proposed method will not be affected by the nonstationary characteristic of BFSAR clutter. In October 2020, the first airborne BFSAR-GMTI experiment in the world has been successfully conducted by us, and the experimental results are given to validate the effectiveness of the proposed method.

Superresolution methods can be applied to real aperture radar (RAR) to improve its angular resolution by solving an inverse problem. However, traditional superresolution methods are achieved after batch data collection, which requires extensive operational complexity and storage space. To solve this problem for RAR, an online detect-before-reconstruct (DBR) framework is proposed in this article based on the sparse property of targets. First, along the range direction, each sample of the echo data is detected to reduce the computational complexity by reducing the dimension of the effective data. Second, along the azimuth direction, a data-adaptive online processing structure is proposed to reduce the storage requirement for the angular superresolution problem. Finally, within the online processing structure, a target data-adaptive updating strategy is proposed to reduce the number of iterations for each target grid. The online DBR-based framework can effectively reduce the operational complexity caused by the noise values of the echo data. Based on the proposed online processing structure, the storage requirement and the operational complexity of the angular superresolution for an RAR system can be greatly reduced without significant reconstruction performance loss. The results of simulations and experimental data verify the proposed framework.

Angular resolution of real aperture radar (RAR) can be improved using deconvolution methods to achieve enhanced target information based on the convolution relationship between target scatterings and an antenna pattern. However, depending on the wide scanning scope and dense sampling angular interval, the computational complexity of the deconvolution methods will drastically increase as the dimension of azimuthal data increases. In this paper, to efficiently improve the angular resolution of RAR, a generalized adaptive asymptotic minimum variance (GAAMV) estimator that relies on a normalized projection array (NPA) model is proposed. On the one hand, the traditional convolution model of RAR is transformed into an NPA model to compress the data dimension. The proposed NPA model can normalize the signal model to make it independent of the sampling parameters. On the other hand, based on the NPA model, a GAAMV estimator is proposed to efficiently reconstruct the targets by adaptively updating each grid. Moreover, the penalty parameter is extended as a generalized case to improve its adaptability to different scenes. Based on the proposed model and method, the computational complexity can be decreased, especially for high-dimensional azimuthal data. Simulations and experimental data verify the proposed model and method.

Marine radar plays a significant role in ship navigation. However, when contending with interference among cosailing navigation radars, the echo data may be unintentionally corrupted, and it becomes challenging to obtain high-quality imagery using current radar imaging methods. To overcome this problem, an efficient anti-interference imaging framework is presented in this article based on the theory of nonuniform sampling. First, a beam-recursive anti-interference method based on the signal-to-interference-plus-noise ratio (SINR) estimation is proposed to compensate for the shortcoming of the traditional interference rejection method. Second, a nonuniform sampling model is established to well model the echo data with missing samples, which facilitates reconstructing the marine radar imagery from the missing echo data. Finally, a fast super-resolution method based on the dimension-reduction iterative adaptive approach (DRIAA) is proposed to reconstruct the distribution of sea-surface targets at a much lower computational complexity. Simulated and experimental results demonstrate that our anti-interference imaging framework can provide radar imagery with higher quality and lower computational complexity than the existing radar imaging methods in the presence of unintentional interference.

The 2-D space variance of echo signals is the key problem of image processing for bistatic synthetic aperture radar (BiSAR) under nonlinear platform trajectories. Although there are some existing studies that present solutions to deal with space variance under linear or low-order motion, this problem is more severe when the trajectories are more complicated and the imaging scene sizes are larger. To deal with this challenging problem, this article proposes a new imaging method based on a novel azimuth-regularized wavenumber mapping and a highly efficient spatially variable phase filter. The novel wavenumber mapping as the major novelty of the method can simultaneously realize range–azimuth decoupling and coarse focusing in the space domain for all the targets. What is more, the phase function of the echo signal in the wavenumber domain is analytically expressed as a binary polynomial by deriving the inverse mapping of the wavenumber with respect to range frequency and azimuth time. Then, the spatially variable filter is designed based on the analytical expression and realized by upgrading the parameters along the azimuth direction. The filtering process has low complexity and can be executed in parallel for every cross-azimuth cell. Moreover, an algorithm for constraining the residual phase error is presented to guarantee both the efficiency and the accuracy of the image processing. Verified by numerical simulations, the proposed imaging method achieves a superior focusing effect than the existing method that is designed for BiSAR with highly maneuvering platforms while having lower computational complexity.

The unmanned aerial vehicle (UAV) is a low-cost and high-efficiency lightweight synthetic aperture radar (SAR)-mounted platform that can be used for a variety of military and civilian missions. Using multiple UAVs to form a swarm can break through the limitations of a single platform and has broad application prospects. In this article, swarm UAV SAR that contains tens or hundreds of UAV platforms is proposed for the first time. The concept and advantages of swarm UAV SAR are investigated, and the mission outlook is given. Afterward, the swarm UAV 3-D linear array SAR (LASAR) is illustrated, which enables high-resolution 3-D imaging in a single flight. Since the antenna array of the swarm UAV 3-D LASAR is sparse, the compressed sensing (CS) algorithm is applied, whose reconstruction performance is closely related to the correlation coefficient of the sensing matrix. Hence, the signal model of swarm UAV 3-D LASAR is derived, and the expression of the sensing matrix is deduced. The sensing matrix design in this article aims at obtaining satisfactory reconstruction performance by optimizing the distribution of the antenna elements, which directly influences the correlation coefficient of the sensing matrix. Considering the limitation of the practical conditions, the sensing matrix design problem is modeled as a constrained integer programming problem. Finally, a sensing matrix design method based on discrete constrained differential evolution (DCDE) algorithm is proposed to solve the optimization problem. Experimental results demonstrate the effectiveness and superiority of the proposed method.

Due to the division of the transceiver, a bistatic synthetic aperture radar (SAR) gains many advantages, such as forward-looking imaging and powerful anti-interference capabilities. In the meantime, information sharing (e.g., positions and status) between the transmitter and the receiver is required for SAR imaging. This article addresses the co- use waveform design for SAR-dual-functional radar communication (SAR-DFRC). To this end, we embed information into a time-frequency spectrum of the phase coded signal, which can be a feasible solution to SAR-DFRC, bringing the possibility of realizing a light-weighted, miniaturized, low-costed, spectrum reusable, and more confidential system. A novel time-frequency spectrum shaping (TFSS) SAR-DFRC architecture based on short-time Fourier transform (STFT) is proposed for the first time. The weighted peak sidelobe level (WPSL) is considered a figure of merit for SAR imaging performance. Information is embedded by nulling the time-frequency spectrum of the waveform. Here, we develop the majorization-minimization PSL-TFSS (MMPSL-TFSS) algorithm to solve the resulting nonconvex NP-hard optimization problem. The designed waveform can ensure imaging performance and obtain high communication capacity in the meantime. Experimental and numerical results highlight the effectiveness of the proposed SAR-DFRC framework for imaging performance and the secure transition to communication information.

Synthetic aperture radar (SAR) automatic target recognition (ATR) is an essential field in SAR application. However, a sufficient number of labeled training SAR images for each target type plays a crucial role in existing SAR ATR methods, while the acquisition and annotation of SAR images are difficult and time-consuming in practice. Therefore, the recognition under the limited labeled training SAR images is the basic and crucial problem in SAR application. In this paper, we propose a novel hierarchically-designed lightweight method (HDLM) by recognition in label and discrimination in feature to address the problem of limited data in SAR ATR. The proposed method is hierarchically designed from top to bottom. In the top phase, the framework is constructed by dual loss to force the deep model to optimize by label recognition and feature discrimination, which is noted as recognition in label and discrimination in feature. In the middle phase, the architecture of the network is built up using a novel lightweight extractor and multi-level cross fusion to boost the amount and diversity of the features for the framework. In the bottom phase, two modules, coordinate attention, and depth-wise separable convolution modules are employed to enhance the feature quality and density with fewer parameters for the phases above. The experimental results on MSTAR and OpenSARship showed that the proposed HDLM performs better than the existing methods under the limited training samples.

Multistatic synthetic aperture radar (SAR) has the characteristics of all-day, all-weather and high-resolution imaging. It can observe a target from different directions simultaneously to obtain multi-angle observation information. However, jamming signals can affect multistatic SAR. When multiple range-deception jammers exist in the environment, multiple false targets are generated in the multistatic SAR image simultaneously, which can impact the readability of the information contained in multistatic SAR images. The locations of false targets are related to the configuration of multistatic SAR, it provides the potential for jamming suppression by adjusting the configuration. Thus, in this paper, we propose a jammer localization and jamming suppression method for multistatic SAR in a multi-jammer environment via configuration optimization. Firstly, a target detection algorithm and a discriminant algorithm are combined to detect and identify false targets. Then, the distribution law of false targets is analyzed, and false targets are classified into two categories according to the types of jammers. Combined with the multistatic SAR configuration and false target information, localization methods for range-deception jammers with different time delays are proposed. Finally, we model the configuration optimization problem as a multi-objective optimization problem (MOP), and the nondominated sorting genetic algorithm II is employed to solve the MOP. As a result, the configuration distribution of multistatic SAR can be altered to exclude false targets from the region of interest, thereby obtaining a multistatic SAR image without false targets. Simulation results demonstrate that the proposed method is effective.

Multistatic SAR is able to observe targets from different angles simultaneously, which enhances the information acquiring capability. However, multistatic SAR can still be affected by electromagnetic jamming, resulting in the misinterpretation of multistatic SAR images. This article proposes a method to locate multiple range-deception jammers and suppress jamming signals. First, the echo model of multistatic SAR under a multijammer environment is established. Second, the detection of interested targets in multistatic SAR images can be achieved through visual saliency detection methods based on spectral residual. Third, location distribution features of false targets in multistatic SAR images are analyzed, and the Euclidean distance criteria are used to effectively distinguish false targets. Accurate localization is then achieved by combing multistatic SAR configuration information. Finally, using a linear constrained minimum variance beamforming algorithm to suppress jamming signals, multistatic SAR images without jamming signals can be obtained. Simulation results validate the effectiveness of the proposed method in this article.

Bistatic synthetic aperture radar (BiSAR) has received more and more attentions because of its forward-looking imaging capability and configuration flexibility. For BiSAR moving target imaging, its non-cooperative motion leads to unknown range cell migration (RCM) and additional phase modulation. Consequently, moving target imaging in BiSAR face two main challenges: 1) The unknown RCM correction and Doppler parameter estimation are tightly coupled. 2) The Doppler parameters of the extended moving target’s different scattering points are different, i.e., the Doppler parameters are spatially variant. To cope with these problems, an optimal polar format refocusing method for bistatic SAR moving target imaging is proposed. First, the main part of tight coupling and spatial variation effects caused by the BiSAR platforms are eliminated, while the moving target is two-dimensional (2-D) defocused and shifted. Then, we analyze the characteristics of two-dimensional defocused and shifted of the moving target in BiSAR, and give the analytical expressions. On this basis, a new bistatic polar format transformation is introduced, in which the degree of freedom of defocusing result is reduced from 2-D to only one-dimension. After that, the parameter estimation and refocusing issues are transformed into a constrained optimization problem (COP), and differential evolution (DE) is applied to solve the COP and obtain the refocusing results. Finally, considering the spatial variation of the extended moving target, the compensation processing is performed to relocate each scattering point. Numerical simulations verify the effectiveness of the proposed method.

Cross-resolution enhancement for airborne high-squint radar (AHSR) imagery is mathematically equivalent to the ill-conditioned problem of inverse-scattering reconstruction. Although a variety of inversion methods with regularization can be introduced to advance the field of AHSR imagery, they turn out to be computationally intensive when extended to 2-D (range and cross-range dimension) image formulation due to the range-by-range calculation for the space-variant inversion operators over the full range swath. To tackle the problem of efficiency, this article presents a low-complexity inverse-scattering strategy. Our underlying idea is to equalize the space-variant Doppler centroid embedded in an inversion operator for a reference range cell using Doppler centroid compensation. With the proposed strategy, the necessary computational complexity required for 2-D AHSR inverse-scattering reconstruction can be significantly reduced by requiring only the calculation of the inversion operator, independently of the number of range cells. Our experimental assessment, conducted using both the simulation and real data, demonstrates that our proposed inverse-scattering strategy offers preferable computational reduction in the task of inverse-scattering reconstruction for 2-D AHSR imagery without resolution loss.

Recently, a variety of super-resolution (SR) methods have been devoted to enhancing the angular resolution of real beam mapping (RBM) imagery in modern microwave remote sensing applications. When addressing large-scale datasets, however, they suffer from notably high computational complexity due to high-dimensional matrix inversion, multiplication, or singular value decomposition (SVD). To overcome this limitation, this article presents a low-complexity SR strategy based on adaptive low-rank approximation (LRA). Our underlying idea is first to construct a random matrix sketching to sample the raw echo measurements and restore the surface map of reflectivity in a low-dimensional linear space. The resulting low-complexity strategy enables substantial computational complexity reduction for a group of SR methods, at the cost of introducing a manually adjusted LRA parameter. Using the Fourier transform-based antenna analysis method, we further reveal that the LRA parameter that ensures support resolution improvement can be determined by a closed-form function of the aperture length, the wavelength, and the field of view, allowing for adaptively and efficiently selecting the optimal LRA parameter that well balances the tradeoff between LRA error and computational efficiency. We use both simulated and real datasets to demonstrate that the proposed LRA-based SR strategy can provide significant speedup without performance loss.

Waveform design has become an attractive topic in the field of colocated multiple-input multiple-output (MIMO) radar that allows antennas to transmit different waveforms. Waveform properties of MIMO radar in space, time and Doppler domains determine the performances of resource utilization, interference suppression, and moving target detection. Therefore, simultaneous optimization of multi-domain properties through waveform design is significant to improve the performance of MIMO radar. In this paper, a novel MIMO radar waveform design framework that constrains the beampattern while maximizing the similarity between the designed and desired waveforms is proposed for simultaneous space-time-Doppler domain optimization. To solve the resulting multi-constraint non-convex problem, an efficient beampattern control and similarity maximization (BCSM) algorithm is developed and its convergence is demonstrated. Especially, the coupling problem due to the similarity constraint is handled by transforming the number domain and introducing the proximal algorithm. While reducing the target distortion in mainlobe region and interference in sidelobe region, the proposed method can also maximize the similarity of MIMO transmit waveforms. Numerical simulation results, apart from verifying that the proposed method outperforms existing methods in space-time-Doppler domain, also illustrate the robustness of proposed method in terms of mainlobe width and desired peak sidelobe level (PSL).

To realize high azimuth resolution for sea-surface targets, this paper proposes a superresolution imaging method that relies on the Weibull distribution. The proposed method introduces the generalized Gaussian distribution and Weibull distribution to represent the statistical distribution function of the target prior information and sea clutter, respectively. The corresponding objective function was derived under the maximum a posteriori (MAP) criterion. To address the nonlinearity of the objective function, this paper adopts the NewtonRaphson iterative method to resolve it. Simulations and experimental data assessment indicate that the proposed method has superior superresolution imaging performance compared with other traditional superresolution methods for sea-surface target imaging.

Video synthetic aperture radar (SAR) have received more and more attention in recent years as it can provide continuous images of the observed area. However, the enormous data generated by the multi-frame images in video SAR brings big challenges to its transmission, storage and processing, especially for small unmanned aerial vehicle (UAV) platform. In this paper, we aim at proposing an efficient video formation method for SAR system with reduced data. The video formation problem is modelled as a joint low-rank and sparse tensors recovery problem. After that, this problem is solved by an efficient tensor recovery method based on alternating direction method of multiplier. Compared with frequency-domain or time-domain imaging methods, the amount of data samples used can be greatly reduced. Numerical simulations validate the effectiveness of the proposed method.

The rapid development of generative adversarial nets (GANs) has led to an increasing number of applications for the synthetic aperture radar (SAR) automatic target recognition (A-TR) with a small sample set in the past few years. However, the generated samples by the GAN s sometimes even lead to a decrease in the performance of the ATR model. In this paper, we propose a filtering approach to address this harm of generated samples. The proposed filtering approach is based on a stable generation model. The stable generation model can continuously and stably generate different batches of target samples. Then, multiple SVMs trained by different SAR target sample sets provide pseudo-labels to the other SVMs to improve the accuracy of the filtering results. Therefore, the proposed approach improves the recognition ability of the A-TR model dynamically while continuously filtering generated target samples. Several experimental results show the superiority of the proposed filtering approach based on the Moving and Stationary Target Acquisition and Recognition (MSTAR) data set. When the number of training samples is 14.5% of the original training set, the recognition rate of the ATR model still reaches 91.27% with the help of the proposed approach.

The echo signal of slow-moving target is usually submerged in clutter signal. As a consequence, ground moving target (GMT) separation is challenging because of the decreasing of detection performance. To solve this problem, this paper proposes a moving target detection method with single-channel for bistatic forward-looking SAR (BFSAR) which mainly contains three steps. First, range cell migration (RCM) is corrected by Keystone transform. Next, Extended Nonlinear Chirp Signal (NLCS) algorithm is applied to equalize the spatial variant Doppler parameters and subpress clutter spectrum broadening. At last, according to the difference of Doppler FM rate between GMT and the stationary clutter, the Short Time Fourier Transform (STFT) is devoted to separate GMT from the stationary clutter. The effectiveness of the detection method proposed in this paper has been demonstrated and illustrated by numerical simulations.

Synthetic Aperture Radar (SAR) image is highly useful in civilian and military fields, such as ship detection, maritime search and rescue. Considering the target detection from SAR image, we propose a SAR image reconstruction and target extraction method via low-rank and sparsity constrains from under-sampled data. Firstly, the low-rank and sparsity constrains are incorporated into the SAR image reconstruction model, and the objective function is established based on Robust Principal Component Analysis (RPCA) theory. Then, the Augment Lagrange Multiplier (ALM) algorithm is used to transform this objective function to a convex optimization problem. Lastly, SAR image reconstruction and target extraction are obtained by Alternating Direction Method of Multipliers (ADMM) algorithm. The simulations are conducted to verify the effectiveness of the proposed method.

Traditional synthetic aperture radar (SAR) imaging methods focus on high-resolution imaging of the observation scene. In the application of specific target imaging and detection, such as threat target search, the priori knowledge of target can be used for better task performance. Therefore, it is highly desirable to improve the image quality of the target. In this paper, we propose a SAR formation method based on the sparse dictionary. Firstly, the sparse dictionary is learned through the SAR images of a specific target via K-SVD method. Then the SAR formation model is established as a sparse reconstruction problem by incorporating the sparse dictionary. Lastly, the problem is solved via Augment Lagrange Multiplier (ALM) method and Alternating Direction Method of Multipliers (ADMM) method. The proposed method is validated by the simulation, and the results show that the proposed method can effectively reconstruct the target.

Convolutional neural networks (CNN) show superior potential in synthetic aperture radar automatic target recognition (SAR ATR). However, due to the difficulty of obtaining SAR images and the scarcity of labeled SAR images, supervised learning has poor performance in this area and is not widely applicable. To address this problem, a semi-supervised conditional generative adversarial network with a multi-discriminator (SCGAN-MD) is proposed in this paper. In our method, a conditional generative adversarial network (CGAN) is adopted with two discriminators for training the generated images and predicting the labels for unlabeled samples. Compared with other semi-supervised learning-based methods, our proposed method has more accurate image generation capability and can achieve improved recognition accuracy of SAR ATR. Experiments on the Moving and Stationary Target Acquisition and Recognition (MSTAR) database indicate that the proposed method can effectively improve the recognition accuracy and robustness of the network with a small number of labeled samples.

The survivability and stability of synthetic aperture radar (SAR) in the increasingly severe electromagnetic environment are widely concerned. With the development of interception receiver technology, it is very difficult to prevent the competitor from detecting the radio frequency (RF) energy of radar. Therefore, more complex intra pulse modulation waveform is needed to prevent the competitor from effectively acquiring and analyzing information. In this paper, a novel low probability of intercept (LPI) waveform optimization framework is proposed. Symmetric piecewise linear functions (PWL) is used to define the waveform search space and a constrained multi-objective evolutionary algorithm is employed to solve the waveform optimization problem. Simulation results show that the proposed waveform has good performance of low interception and auto-correlation, which is suitable for SAR imaging.

Airborne radar high-squint looking imaging is an important research for remote sensing. The traditional Doppler beam sharpening based on fast Fourier transform (FFT) has good real-time performance but low cross-range resolution. Many super-resolution methods have been proposed to enhance the cross-range resolution for airborne radar. However, these methods generally adopt the batch processing mode with high computational complexity and high memory usage, which lead to poor real-time performance. This paper proposes an online super-resolution imaging approach for airborne scanning radar based on sliding window recursive least square (SWRLS) algorithm. The current scattering estimation can be derived recursively through downdating and updating. The proposed method effectively improves the cross-range resolution as well as the real-time performance and memory occupancy, which is beneficial to high-quint continuous realtime imaging for airborne radar. Simulation results are given to demonstrate the effectiveness of the proposed method.

In order to overcome layover and shadowing effects, three-dimentional(3-D) synthetic aperture radar(SAR) systems have developed rapidly. One of them is linear array SAR(LASAR). Therefore, the image result may defocus due to the antenna phase centers(APCs) motion measurement error. Thus a novel LASAR motion compensation approach is proposed. Based on the analysis that linear array motion error is approximately linear, the method estimates only selected APCs' phase error and derive the whole array's phase error via extrapolation. The compensation scheme is verified with simulations.

Video Synthetic Aperture Radar (SAR) has received extensive attention in recent years due to its continuous imaging capabilities. Video SAR requires continuous image reconstruction, and there are many redundant operations in the reconstruction process. In order to meet the real-time requirements of video SAR and improve data utilization, we propose a subaperture imaging algorithm based on PFA. First, the echo is divided into multiple subapertures, and coarse focusing is achieved respectively. Then the subaperture echo is subjected to wavenumber mapping to obtain high-resolution high-frame image output. All subaperture data has only been coarsely focused once, and interpolation is not required for wavenumber mapping, which improves the efficiency of image output. Finally, we use experimental data to verify the effectiveness of the algorithm.

This paper unrolls some preliminary results of a spaceborne-airborne bistatic SAR experiment, conducted in October, 2020 in Zhejiang, China, using GF-3 SAR satellite as the transmitter. Some important aspects of the experiment are firstly introduced, including bistatic acquisition geometry, receiving system, signal synchronization and theoretical spatial resolution. Then, the imaging processing flow is given, with emphasis on the data synchronization. A modified BP imaging method is proposed, which is suitable for direct signal synchronization scheme commonly used in spaceborne-airborne experiments. Finally, the imaging result is given with evaluation of the spatial resolution.

Geosynchronous spaceborne-airborne bistatic synthetic aperture radar (GEO-SA-BiSAR) consists of GEO transmitter and airborne receiver, which has extensive application prospects in both civilian and military fields. However, the Doppler bandwidth in this configuration exceeds the transmitted pulse repeat frequency (PRF), which leads to sub-Nyquist sampling. To solve this problem, multi-receiving technique has been applied to the receiver to increase the equivalent sampling rate and reconstruct unambiguous image. In this paper, we take a different approach to recover the unambiguous image for GEO-SA-BiSAR with less receiving channels. The GEO-SA-BiSAR imaging problem is modeled as a problem of joint sparse and low-rank matrices recovery. To reduce the computing time of the traditional recovery method, a modified alternating direction method of multipliers (M-ADMM) is proposed, where the computing and storage of the computational expensive observation matrix is avoided. Simulation results reveal that the proposed method can recover the original image scene with high computational efficiency.

Unmanned ariel vehicle (UAV) can provide superior flexibility and cost-efficiency for modern radar imaging systems, which is an ideal platform for advanced remote sensing applications. In this paper, an energy-efficient passive UAV SAR system is proposed and investigated. The UAV platform passively reuses the backscattered signal from an external illuminator, such as SAR satellite, GNSS or ground-based stationary commercial illuminators, and achieves data communication and bi-static SAR imaging at a ground processing station. The mission concept and system block diagram are first presented with justifications on the advantages of the system. A set of mission performance evaluators is established to quantitatively assess the capability of the system in a comprehensive manner, including UAV navigation, passive SAR imaging and data communication. Finally, the validity of the proposed performance evaluators are verified by numerical simulations.

The clutter background poses a challenge to the detection and recognition of targets from synthetic aperture radar (SAR) images, especially when the target is submerged by the clutter. In this paper, we propose a target-oriented SAR waveform optimization method to deal with this problem. The proposed method constructs an optimization criterion jointing the signal-to-clutter ratio (SCR) and the resolution of transmitted waveform. Based on the prior information of the frequency response of the interested target, the clutter suppression performance and the range resolution performance are jointly optimized. The simulation results show that the proposed method can effectively improve the SCR of SAR image in the condition of low SCR, while the resolution performance is guaranteed.

Real aperture super-resolution (RAS) technology is widely used in the field of radar forward-looking imaging. However, traditional RAS technology is based on the space-to-ground scanning mode. The essence of this technology is azimuth (angle) super-resolution, which is a one-dimensional super-resolution technology. In our work, we consider applying RAS technology to the space-to-space scanning. In this mode, we regard the echo of each range slice as the convolution of the target scattering coefficient distribution and the antenna pattern function. Its essence is azimuth and pitch super-resolution, which is a two-dimensional super-resolution technology. Finally, a reasonable objective function is constructed under the framework of regularization, and the ADMM solver is used to achieve two-dimensional super-resolution imaging. Simulations will prove the effectiveness of the proposed two-dimensional super-resolution algorithm.

In the field of radar electronic reconnaissance, radar signal recognition is a key technology. In the actual task, part of the signals to be recognized may come from new types of emitters, which can not be identified directly by the existing recognition system. In order to get the ability to recognize new categories, it is necessary to analyze the unrecognized samples for incremental learning. In this paper, a new categories identification method based on reliability test is proposed. Firstly, an existing clustering method is used to label the unrecognized samples, and then the reliability test criteria are designed, including quantity criterion, distance criterion and frequency criterion, to screen the clustered sample points. The proposed method provides better data support for incremental learning in radar signal recognition. Simulation results show the effectiveness of the proposed method.

Multistatic SAR can obtain information from different angles simultaneously, it has application potential in many fields. However, it still be affected by jammers. When deceptive jammers jamming the SAR system, the false target is generated in the SAR image. In this paper, when the false target is generated on the sea by jammer which located at the coast, the anti-jamming method is proposed. Firstly, echo model of multistatic SAR in deceptive jamming environment is established. Then, false targets are detected and recognized. Next, the jammer localization method is proposed according to relationship between the jammer and false targets. Finally, the jammer location is obtained and jamming signal suppression is achieved. Simulation results validate the effectiveness of the proposed method in this paper.

Miniaturized radar near-field imaging systems have become a hot spot for SAR imaging in recent years. The irregular synthetic aperture caused by arbitrary motion of miniaturized system is an important factor affecting imaging quality. On the other hand, the design of fast and high resolution 3D imaging methods is also a major issue. In this paper, a fast time-frequency joint three-dimensional imaging algorithm based on aperture linearization is proposed. First, irregular synthetic aperture is adjusted through special linearization method, and then the target is quickly and accurately imaged. Through formula derivation and simulation test analysis, the results show that the proposed method is feasible and maneuverable, which can solve the current problems.

The echo received by ground-to-air radar is a range-azimuth-pitch three-dimensional data. After pulse compression, the data of each range unit can be regarded as an azimuth-pitch two-dimensional (2D) echo. The resolution of azimuth and pitch is limited to antenna aperture. In this paper, the well-known Wiener filtering, Richardson-Lucy (RL) and total variation (TV) methods are introduced to simultaneously improve the azimuth-pitch resolution of ground-to-air radar. We first analyze the received signal of ground-to-air-radar, and model the echo of each range unit as a 2D convolution of target reflectivity distribution and azimuth-pitch antenna pattern. Then we deduce the Wiener filter, RL and TV methods in detail, and theoretically realize the super-resolution imaging of the azimuth and pitch. Finally, the super-resolution performance of different methods is verified by simulation.

Iterative adaptive approach (IAA) is an effective super-resolution method to improve the resolution of airborne forward-looking radar imaging. Regretfully, the noise sensitivity caused by the non-full rank of matrix lead to the poor performance under low signal-to-noise ratio condition in the forward-looking imaging process. In response to this problem, a regularized IAA method (RIAA) based on singular value decomposition is proposed in this paper which utilizes singular value theory to decompose the autocorrelation matrix in the iteration which is applied to suppress the noise amplification and keep the main information of targets. Compared with conventional IAA method, the proposed method enjoys a preferable noise suppression performance without image quality degradation. Simulations are given to verify the performance gain.

Designing radar waveforms with desired autocorrelation properties is a key point in the development of cognitive radar. To solve the problem of concealing weak targets by strong targets in detection, we consider minimizing the weighted integrated sidelobe level (WISL) metric in frequency domain where the weak targets are located. In order to directly solve the complex non-convex optimization problem, an iteration algorithm based on the general framework of the iterative sequential quartic optimization (ISQO) algorithm that can guarantee fast convergence to a static point is developed. Numerical simulations are provided to assess the effectiveness of the proposed algorithm.

In SAR images, there are objects with large scale difference, which is called cross-scale objects. For example, there are both large-scale airport objects and small-scale ship objects in SAR images. However, the current multi-scale object detection methods are difficult to detect objects with large scale difference. To address this issue, we propose a cross-scale object detection method for SAR images based on Scale Expansion Pyramid Network(SEPN) in this paper. The proposed SEPN can extract the salient features of the objects with a large scale difference, and by closely connecting the scale expansion layer with the convolutional layer during the downsampling process of the Feature Pyramid Network (FPN), the receptive field of the feature extracted by the convolutional layer can be adaptively extended, and finally it achieves the effect of cross-scale object detection in SAR image. Experiments on SSDD dataset and Gaofen-3 dataset show the effectiveness of our proposed methods in detecting objects of different scales in different scenes of SAR images.

A polar format algorithm (PFA) as a high-efficient imaging method is widely used in bistatic spotlight synthetic aperture radar (SAR). In cases where motion error causes great defocusing in PFA images, phase gradient autofocus (PGA) is suitable for motion error compensation due to its robustness. However, a significant coupling of range-azimuth directions in bistatic SAR (BiSAR) makes conventional PGA unapplicable in PFA images. For this limitation, this letter gives the solution to the adaptability of PGA in BiSAR based on the azimuth migration correction in the wavenumber domain. The bistatic motion error is modeled, and the caused phase error is derived and analyzed. Through the wavenumber-domain analysis, the 2-D coupling is indicated still existing after polar format imaging, unlike other algorithms that utilize range cell migration correction (RCMC). Then, a phase compensation for the migration of azimuth direction is proposed to remove the 2-D coupling. Based on this, the new autofocus arithmetic flow for bistatic PFA is designed. The effectiveness of the proposed method is verified by numerical simulations.

Low-earth-orbit (LEO) synthetic aperture radar (SAR) can achieve advanced remote sensing applications benefiting from the large beam coverage and long duration time of interested area provided by a geosynchronous (GEO) SAR illuminator. In addition, the receiving LEO SAR system is also cost-effective because the transmitting module can be omitted. In this article, an imaging method for GEO-LEO bistatic SAR (BiSAR) is proposed. First, the propagation delay characteristics of GEO-LEO BiSAR are studied. It is found that the traditional “stop-and-go” propagation delay assumption is not appropriate due to the long transmitting path and high speed of the LEO SAR receiver. Then, an improved propagation delay model and the corresponding range model for GEO-LEO BiSAR are established to lay the foundation of accurate imaging. After analyzing the sampling characteristics of GEO-LEO BiSAR, it is found that only 12.5% sampling data can be acquired in the azimuth direction. To handle the serious sub-Nyquist sampling problem and achieve good focusing results, an imaging method combined with multireceiving technique and compressed sensing is proposed. The multireceiving observation model is first obtained based on the inverse process of a nonlinear chirp-scaling imaging method, which can handle 2-D space-variant echo. Following that, the imaging problem of GEO-LEO BiSAR is converted to an L1regularization problem. Finally, an effective recovery method named complex approximate message passing (CAMP) is applied to obtain the final nonambiguous image. Simulation results show that the proposed method can suppress eight times Doppler ambiguity and obtain the well-focused image with three receiving channels. With the proposed method, the number of required receiving channels can be greatly reduced.

In synthetic aperture radar (SAR) imaging, moving target is generally mixed with stationary targets. Meanwhile, the image of a moving target is distorted and displaced due to the lack of its prior velocity information. Furthermore, imaging of a moving target for geosynchronous (GEO) spaceborne-airborne bistatic SAR (GEO SA-BiSAR) is a more challenging problem because the echo is sub-Nyquist sampled in azimuth. In this article, a simultaneous moving and stationary target imaging method for GEO SA-BiSAR is proposed. First, range models and the corresponding echo models of moving and stationary targets are established. The observation models for both moving and stationary targets with two receiving channels are derived based on the inverse of an efficient imaging algorithm. After that, the imaging problem of moving and stationary targets is modeled as a joint velocity estimation and sparse decomposition problem, which aims at optimizing the entropy of the moving target image and residual error of the formed images at the same time. Finally, a joint optimization method based on the particle swarm optimization (PSO) method and alternating direction method of multipliers (ADMM) is applied to achieve the imaging of moving and stationary targets and estimation of the moving target velocity. With two receiving channels, the accurate separation and focusing of stationary and moving targets as well as the precise estimation of moving target velocity can be achieved with sub-Nyquist sampling echo. Simulation results are presented to validate the effectiveness of the proposed method.

Ship target detection using large-scale synthetic aperture radar (SAR) images has important application in military and civilian fields. However, ship targets are difficult to distinguish from the surrounding background and many false alarms can occur due to the influence of land area. False alarms always occur with ship target detection because most of the area in large-scale SAR images are treated as background and clutter, and the ship targets are considered unevenly distributing small targets. To address these issues, a ship detection method in large-scale SAR images via CenterNet is proposed in this article. As an anchor-free method, CenterNet defines the target as a point, and the center point of the target is located through key point estimation, which can effectively avoid the missing detection of small targets. At the same time, the spatial shuffle-group enhance (SSE) attention module is introduced into CenterNet. Through SSE, the stronger semantic features are extracted while suppressing some noise to reduce false positives caused by inshore and inland interferences. The experiments on the public SAR-ship-data set show that the proposed method can detect all targets without missed detection with dense-docking targets. For the ship targets in large-scale SAR images from Sentinel 1, the proposed method can also detect targets near the shore and in the sea with high accuracy, which outperforms the methods like faster R-convolutional neural network (CNN), single-shot multibox detector (SSD), you only look once (YOLO), feature pyramid network (FPN), and their variations.

Most existing super-resolution imaging methods fail to work in low signal-to-noise ratio (SNR) condition due to the ill-posed antenna measurement matrix, but the sparse-truncated singular value decomposition (TSVD) method can effectively suppress noise and improve azimuth resolution in low SNR condition. However, the current sparse-TSVD method encounters large computation cost, resulting in a slow algorithm speed. In this work, a fast sparse-TSVD super-resolution imaging method of real aperture radar is proposed. First, the proposed method is based on the results of TSVD, using the truncated unitary matrix and diagonal matrix to reconstruct the signal convolution model. The dimension of the reconstructed antenna measurement matrix reduces from$N \times N$to$k \times N$, and the dimension of the reconstructed echo matrix reduces from$N \times 1$to$k \times 1$, where$N$is azimuth sampling points and$k$is truncation parameter,$N \gg k$. Much of the expensive matrix– multiplication computation can then be performed on the smaller matrices, thereby accelerating the algorithm. Second, an objective function is established as the${l_{1}}$constraint based on the regularization strategy. Lastly, this article employs iterative reweighted least square (IRLS) method to solve the objective function, and the dimension of the reversed matrix is lessened from$N \times N$to$k \times k$, speeding up the algorithm further. The simulation and real data verify that the proposed algorithm not only improves the azimuth resolution in low SNR condition but also increases computational efficiency compared with the sparse-TSVD method.

Bistatic forward-looking spotlight synthetic aperture radar with stationary transmitter and maneuvering receiver (STMR-BFSSAR) is a promising sensor for various applications, such as the automatic navigation and landing of maneuvering vehicles. Because of the bistatic forward-looking configuration and the receiver's maneuvers, conventional image formation algorithms suffer from high computational complexity or small size of a well-focused scene if applied to STMR-BFSSAR. In this article, we propose a wavenumber-domain algorithm for STMR-BFSSAR image formation, which is termed the bistatic-range-Doppler-aperture wavenumber algorithm (BDWA). First, a novel range model in bistatic-range and Doppler-aperture coordinate space instead of conventional Cartesian coordinate space is established by employing the elliptic polar coordinate system and the method of series reversion. The novel range model not only makes the echo's samples to be regular along the direction of the bistatic-range wavenumber axis but also constructs a curved wavefront close to the true wavefront. Second, an operation termed wavenumber-domain gridding is conceived to regularize the echo's samples along the Doppler-aperture wavenumber axis, which can be implemented by 1-D interpolation. The proposed algorithm significantly outperforms the conventional algorithms in terms of computational complexity and scene size limits. Both point and distributed targets are simulated for two STMR-BFSSAR systems with different parameters. The simulation results verify the validity and superiority of the proposed BDWA.

Due to its ability of forming continuous images for a ground scene of interest, the video synthetic aperture radar (SAR) has been studied in recent years. However, as video SAR needs to reconstruct many frames, the data are of enormous amount and the imaging process is of large computational cost, which limits its applications. In this article, we exploit the redundancy property of multiframe video SAR data, which can be modeled as low-rank tensor, and formulate the video SAR imaging process as a low-rank tensor recovery problem, which is solved by an efficient alternating minimization method. We empirically compare the proposed method with several state-of-the-art video SAR imaging algorithms, including the fast back-projection (FBP) method and the compressed sensing (CS)-based method. Experiments on both simulated and real data show that the proposed low-rank tensor-based method requires significantly less amount of data samples while achieving similar or better imaging performance.

An efficient echo simulation algorithm considering topography in time domain for bistatic SAR is proposed in this paper. By using subaperture processing and equivalent scatter, this method can implement efficient echo simulation. At last, the effectiveness of this method is verified by numerical simulation.

This paper presents a long-time integration method for passive radar detection of marine target with multi-stage motions using global navigation satellite system (GNSS) as illuminator. Owing to the complex motion, the range cell migration (RCM) and the Doppler parameters of the target echo pertaining the multiple motion stages are different. To solve the problems, in this method, first the range symmetry transform (RST) is applied to correct the RCM over the multiple stages, after which a one-dimensional azimuth signal can be directly extracted from the two-dimensional echo. Then, the conjugate integration processing (CIP) is performed to accumulate the target energy during each stage into Doppler centroid (DC) and Doppler frequency rate (DFR) domain. Finally, in order to settle the difference in Doppler parameters between multiple stages, the DC-DFR maps are projected and combined in high-dimensional Doppler parameter domain after Doppler frequency shift compensation, implementing the well integration of the echo over the entire long time. The effectiveness of the proposed integration method is demonstrated by simulations.

In this paper, an improved target extraction scheme is proposed for forward-looking scanning radar. In the scheme, an image with good azimuth resolution is obtained firstly by deconvolution technique, and then it is used to map a patch-image. Secondly, low rank and sparse matrix decomposition is conducted on the patch image. Thirdly, constant false-alarm rate (CFAR) detection method is used to extract the targets from the sparse matrix. Finally, simulation results are given to verify the effectiveness of the proposed scheme.

For a non-cooperative ground moving target (GMT), its complex motion inevitably induces unknown range cell migration (RCM) and Doppler frequency migration (DFM) in Bistatic SAR (BiSAR) echo. Unfortunately, the existence of RCM and DFM often results in a deteriorative or even unacceptable performance on target detection and parameter estimation. In this paper, a novel coherent integration approach for BiS-AR GMT detection and parameter estimation based on two-dimensional (2-D) deramp processing is proposed. First, the deramp processing along the range frequency is exploited for GMT's unknown RCM correction. Then, the second deramp function along the slow time is constructed to reduce the order of azimuth phase and eliminate the effect of DFM. Finally, GMT's energy can be accumulated into a peak and Doppler parameter can be obtained via azimuth fast Fourier transform (FFT) and range inverse FFT. The proposed approach is computationally efficient, since it can be implemented only by complex multiplications and FFT. Simulations are given to verify the effectiveness of the proposed approach.

Radar antenna can work in scanning mode to obtain a wide region observation. However, for the targets located at the scene edge, the targets are only swept by less than half of the radar beam. Therefore, the scene edge targets are recovered distortedly using the conventional angular super-resolution methods. To keep the performance of recovered targets in the full scene, in this paper, a data extrapolation-based parallel iterative adaptive approach (PIAA) is proposed. First, we analyze the cause of scene edge target distortion. Then, the echo data is extrapolated by half of the radar beam to compensate the unobserved data. Last, a parallel iterative adaptive approach is proposed to recover the targets efficiently. Simulation data is applied to verify the proposed method.

Synthetic aperture radar (SAR) has always received wide attention for its developing performance in military and civil applications. SAR automatic target recognition (ATR) is an important research field of the SAR application with the growing number and resolution of the SAR images. SAR images will be greatly influenced by the imaging azimuth, which could also be utilized to extract the correlation features between the adjacent azimuths. In this paper, we proposed a multi-view convolutional neural network and long short term memory (CNN-LSTM) network to extract and fuse the feature extracted from different adjacent azimuths. It adopts the structure of convolutional neural network to extract the optimal feature from the SAR images. Then, the structure of multiple layers of the long short term memory is adopted to fuse the optimal features of adjacent azimuths. Finally, a softmax is employed as the classifier to get the recognition results. Experimental results based on the MSTAR data set have shown the effectiveness and accuracy of the proposed method.

In SAR image target detection, harbor detection can help the detection of harbor targets and maritime traffic planning. In this paper, we propose a harbor detection method of SAR images based on multidirectional one-dimensional scanning. Take the candidate points along the coastline and the multidirectional one-dimensional scanning is performed. Using the distribution characteristics of land, sea and dock in the one-dimensional vector, training a convolutional neural network to classify the candidate points into harbor and non-harbor feature points. Then we get the harbor feature points map reflecting the distribution of harbors. The Sentinel-1 spaceborne SAR images covering a coastal region are used to verify the proposed method. The experimental results show the effectiveness and accuracy of the proposed method.

Synthetic Aperture Radar (SAR) is widely used in environmental monitoring and disaster early warning due to its high resolution imaging performance. A distributed radar system can be established by mounting radars on multiple unmanned aerial vehicle (UAV) platforms. Distributed radar utilizes multiple transmitters distributed in different spatial positions, flying along a certain planned path and enable multiple transmitters to obtain as large an aperture as possible in a certain time. In this paper, an intelligent optimal path planning method for distributed radar short-time aperture synthesis is proposed, which can deal with terrain obstacles and line-of-sight occlusion in UAV flight path and achieve the goal of maximum aperture accumulation in a specific time. Simulation results verified the effectiveness of the UAV intelligent optimal path planning method.

In this paper, a weighted dense connected convolutional network(WDCCN) is proposed for SAR image super-resolution. In the network, to enhance feature propagation and the super-resolution performance, each layer will receive the output from all the previous layers in a different weight proportion. At last, the experimental results indicate that weighted dense connected convolutional network can realize SAR image super-resolution well.

Sparse regularization method has been widely used to realize super-resolution imaging in radar forward-looking imaging. However, most of existed methods directly minimize a nondifferentiable L1 regularization problem. In this paper, a Majorize-Minimization (MM) based super-resolution method is proposed to realize super-resolution for radar forward-looking imaging. According to MM principle, the proposed method converts the non-differentiable L1 regularization problem into a differentiable L2 regularization problem, and the real target distribution is obtained by solving the L2 regularization problem. Due to the introduction of the sparse prior, the proposed method can better improve the azimuth resolution of radar forward-looking imaging. In addition, the application of MM principle makes the non-differentiable L1 regularization easier to be solved. Finally, the superior performance of the proposed method is verified by simulation.

Total variation (TV) method has been utilized to realize super-resolution and preserve contour information of target in radar forward-looking imaging. However, its real-time ability is restricted to matrix inversion. In this paper, a fast TV (FTV) superresolution method is proposed to improve the real-time superresolution ability of traditional TV method. The proposed FTV method utilizes the low displacement rank features of Toplitz matrix and realizes fast matrix inversion by Gohberg-Semencul (GS) representation. It not only effectively improves the azimuth resolution and preserve the contour information of target, but also reduced the computational complexity of traditional TV method to improve its real-time superresolution ability. The superior performance of the proposed FTV method is verified by simulation and measured data processing.

Echoes of bistatic forward-looking synthetic aperture radar (BFSAR) disperse to multiple range cells and exist spatial frequency extension as well as Doppler spectrum extension (i.e., namely space-time extension). Furthermore, the characteristics of BFSAR clutter are strongly nonstationary and spatial variants. Because of the abovementioned issues, clutter and moving targets are fully overlapped in the initial 3-D space-time-range raw data domain, and the clutter cannot be suppressed effectively. To solve this problem, a BFSAR multipulse displaced phase center antenna (MP-DPCA) method is proposed in this article. First, keystone transform without Doppler ambiguity is applied to remove the coupling between the space-time and range domains. Hence, the overall complex 3-D processing in the space-time-range domain is reduced to independent 2-D processing in each space-time domain. Subsequently, a spatial-dechirp processing is applied in the space-time domain to eliminate the spatial frequency extension. Meanwhile, Doppler parameters of clutter point scatterers are equalized by nonlinear chirp scaling processing in the frequency and time domains. Accordingly, the Doppler spectrum extension of point scatterers can be eliminated by a uniform azimuth dechirp processing. After aforesaid three steps, the clutter and moving targets are separated in the space-time domain. Finally, based on the linear phase difference of clutter between the channels, a multipulse canceller can be designed to suppress the clutter. Compared with the existing space-time adaptive procession (STAP) and DPCA methods, this method not only overcomes the nonstationary problem in BFSAR but also conquers the strict application conditions of DPCA. Simulation results are given to verify the effectiveness of the proposed method.

Forward-looking imaging and target detection are highly desirable in many military and civilian fields, such as search and rescue, sea surface surveillance, airport surveillance, and guidance. However, there is a blind zone of forward-looking imaging for conventional Doppler beam sharpening and synthetic aperture radar. Scanning radar can be utilized to obtain a real beam image of a forward-looking area and implement target detection, while its azimuth resolution is poor due to the limitation of antenna size. Besides, during the processing procedure, imaging and target detection are usually regarded as two independent parts, which means that the imaging result will directly affect the detection performance. In this article, an integrated algorithm of super-resolution imaging and target detection for forward-looking scanning radar is proposed. In this algorithm, first of all, low-rank and sparse constraints as regularization norms are incorporated into the forward-looking scanning radar imaging and the objective function is established. Subsequently, the convex theory is utilized to solve the objective function and transform the problem of simultaneous super-resolution imaging and target detection into an optimization problem. Lastly, the super-resolution imaging and the target detection results are obtained simultaneously by solving the optimization problem using the alternating direction method of multipliers. In addition, simulation and experiment results are given to verify the effectiveness of the proposed algorithm.

Because of the poor performance of the conventional total variation (TV) super-resolution imaging method in low signal-to-noise ratio (SNR) condition, a TV super-resolution imaging method based on the truncated singular value decomposition (TSVD) strategy is proposed. First, based on the regularization theory, the TV function is selected as the constraint term to construct objective function. Second, to solve the problem of noise amplification faced by the conventional TV method, this article reconstructs the objective function based on the TSVD strategy, which improves the antinoise performance by discarding small singular values of antenna convolution matrix. Finally, due to the nondifferentiable property of reconstructed objective function, this article utilizes the iterative reweighted norm (IRN) method. Since the influence of the noise is weakened by the TSVD strategy, the proposed method can achieve super-resolution imaging and contour preservation in low SNR condition. The simulation and experimental results demonstrate the effectiveness of the proposed method.

Real-aperture radar can be utilized to realize forward-looking imaging by antenna scanning the imaging region. However, low azimuth resolution seriously affects its practical application. Although traditional super-resolution methods could enhance azimuth resolution to a certain extent, effective preservation of contour information for important targets still remains to be a problem. In this article, a method of total variation-sparse (TV-sparse) multiconstraint deconvolution is proposed to improve azimuth resolution of forward-looking imaging as well as preserve contour information of important targets. Since our interested targets usually appear to be sparse, the sparse constraint of the target is introduced first to achieve high resolution of forward-looking images, which may cause the loss of target contour information in the meantime. Second, total variation (TV) constraint is introduced based on the sparse constraint, converting traditional single-constraint super-resolution problem to a multiconstraint problem. We then use the split Bregman algorithm (SBA) to solve the multiconstraint problem, whose solution is the super-resolution image of radar forward-looking region. Compared with traditional super-resolution methods, the proposed method can improve the azimuth resolution of radar forward-looking imaging as well as better restore target contour information by adjusting respective weights of sparse constraint and TV constraint. Finally, the performance of the proposed method is validated with the simulation and measured data.

This paper briefly introduces the basic scheme of a maritime moving target detection and localization method with multiple BeiDou satellites as transmitters. To detect the target and localize it, a receiver local-coordinates is constituted, whose axes are X, Y, and Doppler frequency rate (DFR), and the received echoes from multiple satellites are integrated into the local-coordinates. In this method, first symmetrical keystone transform (SKT) is applied to correct range cell migration (RCM) and set Doppler centroid (DC) to zero. Then, dechirp-FFT is applied to integrate each echo in fast-time and DFR domain. Finally, each DFR integration result is back projected to local-coordinates according to its bistatic range. From the final integration result, target can be detected and localized simultaneously. The effectiveness of the proposed method is demonstrated by simulations.

Time and frequency synchronization is a crucial problem for Bistatic SAR systems. In this paper, frequency reference error of bistatic SAR system is analyzed firstly. Then the time and frequency synchronization errors are deduced and the intrinsic relationship between them is revealed. At last the effect of the frequency reference error is studied. Simulation and experiment results are given to verify the effectiveness of the analysis.

This paper proposes a clutter suppression method for the bistatic forward-looking SAR(BFSAR). The proposed method relies on a proper processing of the received echoes. First, keystone transform (KT) and the higher-order RCM correction are applied to correct the range cell migration (RCM). Then, azimuth nonlinear chirp scaling (NLCS) and azimuth dechirp are applied to suppress the spatial-variant Doppler spectrum extension. After the aforesaid two steps, the moving targets and clutter can be separated in the space-time domain, and the clutter signal with different pulses contains only one constant phase difference. Thus, the two-pulse clutter canceller can be designed to suppress the clutter. Finally, it is suggested that this proposed method can improve the signal to clutter ratio (SCR) and are beneficial for the subsequent signal processing, e.g. target detection. Simulation results are provided to validate the effectiveness of the clutter cancellation method.

In this paper, an accelerated maximum a posteriori (AMAP) algorithm is proposed to realize fast and effective super resolution imaging of real beam scanning radar. The main idea of this algorithm is to construct a prediction vector based on the first and the second order of difference information before iteration. By using Taylor expansion series and second-order vector extrapolation technique, it aims to enhance the convergence speed of maximum a posteriori algorithm. Finally, the proposed algorithm is verified by simulations.

This paper proposes a parking space information monitoring system by millimeter wave synthetic aperture radar (SAR) based on unmanned aerial vehicle (UAV). Parking space information that people are concerned about includes vacant parking place, parking place occupied by obstacles and place parked by vehicles. Specially, the free parking space detection is an important module for the parking guidance system (PGS) that can help drivers to find parking space efficiently. In this system, we obtain high resolution SAR images of parking lots at first. Then, in order to define the free parking space, Maximally Stable Extremal Region (MSER) method is exploited to leach the candidate regions occupied by vehicles from millimeter wave SAR images. Next, the system utilize visual saliency detection method to extract obstacles from the non-parked parking space acquired by pre-detection. Ultimately, the three types of information have been determined, including vacant parking space, parking space occupied by obstacles and the parked place. Experimental results prove that the integrated scheme performs well in parking information determination.

Target detection is an important function of forward-looking scanning radar search and tracking applications. However, it is difficult to detect the targets using the real beam image with low azimuth resolution. In this paper, a target detection scheme is proposed for forward-looking scanning radar. First, an image with better resolution is obtained by deconvolution technique, and it is used to map a patch-image. Then, according to the low rank characteristic of the patch-image and the sparse characteristic of the targets, the target detection is converted into an optimization problem of low rank and sparse matrix decomposition. Finally, the targets are obtained by solving this optimization problem. Simulations are given to verify its effectiveness.

Scene surveillance radar, which generates radar stochastic radiation field with time and space to obtain more observation information, plays a significant role in disaster monitoring and environmental security. To explore its three-dimensional (3-D) imaging capabilities, in this paper, we propose an echo rearrangement super-resolution imaging method to achieve 3D high resolution imaging for SRR. Because the echo of SRR is uncorrelated along sampling time, we adjust the conventional intrapulse frequency hopping to interpulse frequency hopping. In this way, the proposed method can improve the imaging resolution by echo rearrangement utilizing the noncorrelation with time of stochastic radiation field. The 3-D image provides the scene reflectivity estimation along polar coordinate system including pitch, azimuth and space distance. Simulation results are given to illustrate the performance of the proposed method.

Airborne bistatic (and multistatic) radar, which utilizes the spatial diversity of radar platforms to achieve high-resolution imaging, plays a significant role for the next generation radar. Based on the distribution structure of radar platforms, we can deduce the spatial spectrum to reconstruct the targets. However, the processing efficiency of spatial spectrum is different because the echo data can be projected into different shapes in spatial spectrum region. In this paper, an efficient imaging method based on spatial spectrum projection algorithm (SS-PA) for bistatic radar is proposed. First, the spatial spectrum distribution is illustrated based on the system structure. Then, the minimum external rectangular grids are depicted when the spatial spectrum is projected into different directions. Finally, the most efficient imaging view is obtained according to the ratio of spatial spectrum distribution to processing area. The proposed method can provide the most efficient imaging view for bistatic radar, which reduces computational complexity for the system implementation. Simulation result verifies the proposed method.

Resolution features of spotlight SAR appear space-variant in most cases, however different in PFA. In this paper, we make analysis on the resolution features of PFA and BPA through wavenumber spectrum and draw some novel conclusions. Distinguishing from BPA, the range-Doppler contours appear parallel in the imaging domain of PFA, which makes the space-variation of the resolution features disappear. As verified by simulations, these unique phenomenon will be useful in further research on spotlight SAR.

Multistatic synthetic aperture radar (SAR) have achieved significant performance in the field of anti-jamming for its flexible configuration. Hence, its capacity of anti-jamming is very important in electronic warfare. In addition, it is difficult to remove the false targets caused by the deceptive jammer from SAR echoes. We propose a novel anti-deceptive jamming method for multistatic SAR in this paper which is able to locate the deceptive jammer and eliminate its influence. We first apply maximally stable extremal region (MSER) and European-distances-based method to detect the false targets. The position of the jammer can then be obtained by exploiting the position information of the false targets and the SAR transmitters/receivers. At last, beamforming is applied to achieve the anti-deceptive jamming. Experimental results demonstrate the effectiveness of our proposed method.

SAR ship detection is essential for marine monitoring. Due to the high similarity between the harbor and the ship body on gray and texture features, the traditional methods are unable to achieve effective inshore ship detection. An improved Faster R-CNN based on MSER decision criterion for SAR ship detection in harbor is proposed in this paper. It is a ship detection method based on the combination of feature-based method and pixel-based method. Firstly, Faster R-CNN is used to generate region proposals. Then, replace the threshold decision criterion of Faster R-CNN with the maximum stability extremal region (MSER) method to reassess the generated region proposals with higher scores, aiming at improving the detection rate and reducing the false alarm rate simultaneously. Experimental results based on satellite-borne SAR data illustrate that the proposed method obtains excellent detection performance and low false alarm rate.

High resolution is always the most concerned issue of radar imaging. Traditional radar systems, which obtain echo data using single platform, can achieve limited imaging resolution in a specific view angle. Distributed radar system, which expands multi-platform in space to obtain high imaging resolution by forming a large aperture, is a novel and hot research point. Matched filter, such as inverse fast Fourier transform (IFFT), is a conventional method to deal with distributed radar imaging. However, the method relies strictly on geometric configuration. In this paper, an iterative adaptive approach (IAA) based method is proposed to solve the problem of configuration adaptability. It can maintain the performance of matrix during the iteration. Then, the distributed radar system can keep high resolution in different geometric configurations. Simulation results verified the excellent performance of the proposed IAA-based imaging method.

Distributed radar imaging utilizes expanded array elements in space to form a large aperture and obtain high imaging resolution. A great number of array elements are required in traditional distributed radar system which uses multiple platforms. The distribution of spatial spectrum is affected by the number and the signal form of array elements. In this research, to improve the utilization efficiency of platform resources, a resource allocation optimization method based on Unmanned Aerial Vehicle(UAV) is proposed. It chooses the optimized bandwidth and sampling frequency points of array elements by analyzing the relationship between spatial spectrum and imaging performance. This method can use a small number of UAVs to maintain high imaging resolution. Simulation results verified the effectiveness of the resource allocation optimization method for image quality improvement.

With appropriate geometry configurations, bistatic synthetic aperture radar (SAR) can break through the limitations of monostatic SAR on forward-looking imaging. Thanks to such a capability, bistatic forward-looking SAR (BFSAR) has extensive potential applications. In BFSAR, the compensation of the spatially variant motion errors is of great significance to get a well-focused image. In this paper, a motion compensation method based on keystone transform and modified autofocus back-projection is presented to deal with this problem. Keystone transform is applied to remove the spatial variation of range cell migration (RCM) and the first-order term of RCM errors simultaneously, prepares for the following modified autofocus back-projection, which can eliminate the high-order term of azimuth phase errors. Simulation results verify the validity and efficiency of the presented method.

Bistatic forward-looking synthetic aperture radar (BFSAR) is increasingly in the focus of study, as it breaks through the limitations of imaging on forward-looking terrain of a moving platform. Fast factorized Back-projection (FFBP) is a reliable BFSAR imaging algorithm with both imaging precision and efficiency being taken into consideration. In FFBP imaging, compensation of the motion errors is important to obtain a well-focused image. To accomplish an accurate motion compensation in image processing, a high-precision navigation system is needed. However, in many cases, because of the accuracy limit of such systems, motion errors are difficult to be compensated correctly, resulting chiefly in resolution decrease in the final images. To deal with such a problem, we propose an FFBP imaging algorithm integrated with motion error estimation for BFSAR. We empirically compare the proposed method with several state-of-art BFSAR imaging and autofocus algorithms. Simulations on BFSAR data show that the proposed method is more accurate and has similar computational cost.

The clear contour is required when realize azimuth superresolution of forward-looking radar imaging in many applications. Traditional deconvolution methods achieve the azimuth superresolution but are limited in contour recovery. Although the total variation (TV) method can be used to keep the contour information, it's sensitive to noise because of derivation. In this paper, we propose an improved total variation (ITV) method to realize azimuth superresolution of forward-looking radar imaging and recover the contour information. Firstly, the TV norm and L2norm are combined as the penalties under regularization framework. Then the regularization problem is solved by split Bregman algorithm. The proposed ITV method achieves higher azimuth resolution and better contour recovery performance than traditional methods, and the super performance is verified by simulations lastly.

Regularization technology can be utilized to improve the azimuth resolution for forward-looking scanning radar. In this paper, low-rank and sparse constraints as regularization norms are incorporated into the forward-looking scanning radar imaging. This method can achieve azimuth superresolution and noise suppression. Simulations are given to verify the effectiveness of the method.

Angular resolution of scanning radar is constrained by the size of antenna aperture. Such coarse resolution can not satisfy the applications of microwave remote sensing that require high resolution. Iterative adaptive approach (IAA) is a recently introduced method for scanning radar angular super-resolution, which could notably improve the angular resolution and suppress the noise amplification. In this paper, we further this development, by presenting a beam-recursive I-AA, allowing for adjusting the regularization parameter adaptively and dynamically for varying scenario. Such implementation could effectively eliminate the artifacts on background when applying the batch IAA to resolve closely spaced strong targets. Moreover, the technique offers a promising potential that deserves further attention on computationally efficient implementation and real-time imaging along antenna beam scanning. Simulations are provided to validate the effectiveness of the proposed approach.

Coastal surveillance has long been paid a lot of attention for the threat of flooding due to some natural phenomena, such as global warming. Prompt and accurate reaction to the visualization of the flooded areas is the key. An image fusion rule is thus proposed in this paper to achieve image enhancement of the flooded areas. The rule, targeted at high-frequency parts of the synthetic aperture radar (SAR) and optical images, is able to exploit and combine the merits of both SAR and optical images to obtain the exact flooded areas with the clear boundaries. Experimental results validate the performance of the proposed fusion rule and show that not only the clarity of fusion images is improved, but also the texture and brightness contrast are greatly enhanced.

Geosynchronous (GEO) spaceborne-airborne bistatic synthetic aperture radar (GEO-BiSAR), where the high-altitude transmitter provides continuous illumination for the receiver, is capable of providing benefits to remote sensing applications. However, obtaining a focused image with high efficiency is a challenging work. The severe 2-D space-variant range cell migration and Doppler modulation introduced by the two moving platforms make the echo hard to be focused. Moreover, the azimuth ambiguity due to the low pulse repetition frequency adopted by the GEO illuminator seriously deteriorates the quality of the final image. In order to simultaneously suppress the azimuth ambiguity and obtain well-focused images, a weighted fast factorized backprojection (FFBP) method is proposed for multichannel GEO BiSAR in this letter. First, the Doppler ambiguity of GEO BiSAR is analyzed and the azimuth multichannel receiving technique is introduced. Then, the multichannel transfer function for GEO BiSAR is derived. Based on the multichannel transfer function, a weighted FFBP method is proposed to achieve accurate focusing and ambiguity suppression. Finally, the simulation results validate the effectiveness of the proposed method.

An echo model plays a pivotal role in the processing of synthetic aperture radar (SAR) data. It has been shown that the widely adopted echo models based on stop-and-go approximation (SAG) are unsuitable for use in many emerging radar systems with high resolution or high-speed platforms. More accurate echo models without SAG have been proposed for some special monostatic scenarios, such as low-Earth-orbit SAR and geosynchronous SAR. However, due to their inherent assumptions, such as linear and constant-speed/orbit-based motion, these conventional echo models cannot be applied to general cases, which involve bistatic configurations and maneuvers. Maneuvers or nonlinear/inconstant-speed motion are very common in cases, such as circular SAR and maneuvering-platform-borne SAR. In this letter, we propose a generalized echo model for bistatic SAR, which accounts for maneuvers of the platforms during pulse propagation. Moreover, we obtain a high-precision closed-form expression of the model by approximating high-order terms of the accurate range history of the receiver. Compared with the conventional models, the proposed echo model can be applied to a much greater number of cases, ranging from slow and constant-speed platforms to fast and maneuvering platforms. Theoretical analysis and backprojection-based image formation simulations confirm the validity of the proposed echo model.

Due to the high-altitude illumination and the separation of the receiver and transmitter, geosynchronous (GEO) spaceborne-airborne bistatic synthetic aperture radar (BiSAR) is more flexible and accessible in remote sensing applications. In this paper, the Doppler characteristics of GEO BiSAR with a squint receiver are analyzed. It is found that the Doppler spectrum is generally aliased in GEO BiSAR regarding the low pulse repetition frequency (PRF) adopted by the GEO SAR. In order to suppress the ambiguity without adjusting the PRF of GEO SAR, the azimuth multichannel receiving technique is applied to the receiver and then the multichannel transfer function for GEO BiSAR is derived. However, the whole bandwidth of the imaging scene is much larger than that of the center point, which requires extra receiving channels to suppress the ambiguity and thereby increasing the system complexity. A two-step nonlinear chirp scaling (NLCS) method is proposed to obtain the well-focused image with reduced receiving channels. First, a preprocessing step is conducted to achieve space-variant range cell migration correction. After that, the first-step NLCS processing is applied to equalize the 2-D space-variant Doppler centroid and thereby the Doppler bandwidth is decreased, i.e., the required number of receiving channels for reconstruction is reduced. Then, the unambiguous spectrum is reconstructed based on the proposed multichannel transfer function. Finally, the second-step NLCS processing is carried out to equalize the 2-D space-variant high-order Doppler parameters and obtain the well-focused image. The simulation results validate the effectiveness of the proposed method. With the proposed two-step NLCS method, the well-focused image for GEO BiSAR is obtained and the required number of receiving channels can be decreased, which is beneficial to reducing the system complexity and hardware cost.

Due to its flexibility and low cost, the bistatic forward-looking synthetic aperture radar (BFSAR) which employs side-looking transmitter and forward-looking receiver has been studied in recent years. Sparsity-based techniques have been applied in the field of BFSAR imaging and show great potential. In sparsity-based BFSAR imaging, compensation of the motion errors is crucial to get a well-focused image. For fields that admit a sparse representation, we propose a sparsity-based imaging approach integrated with motion error estimation and compensation in this paper. First, a novel joint phase-amplitude compensation-based motion error correction scheme is developed to cope with the spatial variance of motion error. Then, an inversion observation model of the range-Doppler algorithm combined with motion error correction is derived, based on which a joint problem of BFSAR imaging and motion error estimation is formulated as a sparse recovery problem and solved in an iterative way, where in each iteration, both image formation and motion error correction are carried out. Experiments on both the simulated and real BFSAR data show that the proposed method can obtain a more accurate estimation result, and generate better focused images compared with the existing methods.

Back-projection (BP) is a reliable synthetic aperture radar (SAR) imaging algorithm because of its high-resolution and strong adaptability. However, it is hard to implement because of its high computational complexity. Fast factorized BP (FFBP) is a new way to fix this problem. Like traditional BP, FFBP is compatible with arbitrary flight paths if the track deviations are measured within fractions of a wavelength. However, when the motion information is not accurate enough, autofocus become an important way to get well-focused images. In this paper, we present an effective autofocus method for FFBP to solve the imaging problem caused by platform's motion errors. First, an image quality evaluation function with unknown phase error based on image sharpness for FFBP is established. Then, the phase error computation for autofocus is modeled as an optimization problem. Second, the coordinate descent (CD) and secant processing are introduced to the maximum image sharpness problem. The proposed method keeps the rapid imaging performance of FFBP and solves well the motion error compensation problem. In the end, simulated data and real data were used to verify the effectiveness of the proposed algorithm.

In geosynchronous spaceborne-airborne bistatic synthetic aperture radar (GEO-BiSAR) system, the airborne platform achieves high-resolution imaging by passively receiving the signal from the interested scenario. In this paper, the Doppler characteristics of GEO-BiSAR and the individual contribution of the transmitter and the receiver are first analyzed. The airborne receiver is found to be the dominant contributor for the total Doppler bandwidth, which will lead to Doppler spectrum aliasing regarding the low pulse repetition frequency (PRF) adopted by the GEO-SAR. In order to suppress the Doppler ambiguity without adjusting the PRF of GEO-SAR, azimuth multichannel receiving technique is introduced to the airborne receiver. The multichannel transfer function is derived based on the method of series reversion and the spectrum reconstruction algorithm is then modified for multichannel GEO-BiSAR. Moreover, the reconstruction performance is closely related to the corresponding spacing between each channel (i.e., channel configuration). Therefore, the channel configuration design for GEO-BiSAR aims at optimizing the azimuth ambiguity-to-signal ratio with a satisfactory level of signal-to-noise ratio scaling factor by adjusting the channel configuration. The channel configuration design is modeled as a constrained single objective optimization problem (CSOP). Then, a channel configuration design method based on differential evolution and feasibility rule is proposed to solve the CSOP and obtain the channel configuration for the receiver with the optimal reconstruction performance. Finally, simulations results are presented to verify the effectiveness of the proposed method, and characteristics of channel configuration are analyzed in detail, which can be a practical guide for the implementation of multichannel GEO-BiSAR systems.

Being capable of providing weather-independent, day-and-night, forward-looking, and high-resolution images, bistatic forward-looking synthetic aperture radar (BFSAR) is a promising sensing technique in applications such as the scene-matching-aided navigation for recently emerging high-speed maneuvering platforms (HMPs). Because of the high speed and the great maneuverability of HMPs and the bistatic forward-looking configuration, conventional image formation algorithms, such as polar format algorithm (PFA), are no longer suitable for HMP-borne BFSAR (HMP-BFSAR). Hence, in this paper, we propose a novel PFA for HMP-BFSAR image formation. In the proposed PFA, a range model, termed as quasi-continuous -move range model, is established by taking the maneuvers of the receiver during pulse propagation into account instead of adopting stop-and-go approximation. Moreover, to take advantage of the collected k -set efficiently, an affine mapping, termed as k -set affine mapping, is conceived to transform the parallelogram-shaped k -set support region to a horizontal and quasi-rectangular one. Furthermore, to compensate for the defocus effect induced by wavefront curvature, a closed-form refocus filter based on the implicit function theorem is derived. Both point target simulation and distributed target simulation are presented in this paper. The simulation results show that the proposed PFA significantly outperforms the conventional PFA in terms of focusing quality and computational efficiency when applied to HMP-BFSAR image formation.

Due to the high altitude illuminator and the separation of the receivers and transmitter, Geosynchronous (GEO) spaceborne-airborne bistatic SAR (GEO BiSAR) is more flexible and accessible in remote sensing applications. However, by introducing a high speed airborne platform as receiver, azimuth spectrum aliasing occurs. In order to suppress the azimuth ambiguity without increasing the PRF of GEO SAR system, azimuth multichannel receiving technique is introduced to the airborne receiver in this paper. Firstly, the Doppler characteristics of GEO BiSAR are analyzed. Then, the multichannel transfer function for multichannel GEO BiSAR is derived and a modified multichannel reconstruction method is proposed to suppression the azimuth ambiguity. Finally, simulation results validate the effectiveness of the proposed method.

Geosynchronous (GEO) spaceborne-airborne multistatic SAR (GEO MulSAR) is flexible and accessible in remote sensing applications. Moreover, the information obtained by the multiple airborne receivers can be fused to enhance the spatial resolution. However, the fused spatial resolution significantly depends on the applied multistatic topology. In order to achieve the optimal fused spatial resolution by properly adjusting the imaging topology, a topology design method is proposed in this paper. Firstly, the spatial resolution model of GEO MulSAR is given and the dependance of the spatial resolution on the multistatic topology is analyzed. Then, a topology design method is proposed to obtain the best multistatic topology based on multiobjective optimization methods. Finally, the simulation results validate the effectiveness of the proposed method.

Due to the difficulty of sea scene synthetic aperture radar (SAR) trial, SAR image simulation for sea-ship scene is vitally important for the research of sea remote sensing and surveillance. In this paper, a new SAR image simulation method for sea-ship scene is proposed. Firstly, the geometrical models of sea surface and ship target are obtained through sea spectrum and CAD modeling technology respectively. Then the SAR image intensity data of sea surface is calculated by small perturbation method (SPM) and velocity bunching (VB) theory, meanwhile the radar cross section (RCS) data of ship target is computed through physical optics (PO) method. Finally, the SAR image of sea-ship scene is generated by SAR imaging method after transforming image intensity data and RCS data to the same spectrum domain. The simulation result has verified the effectiveness of the proposed method.

Moving targets are usually defocused in synthetic aperture radar (SAR) images due to across range unit (ARU) range migration and Doppler migration caused by unknown motion parameters. To eliminate the influence of these issues on moving target imaging, a novel Doppler parameters estimation method for moving targets is proposed. First, second-order keystone transform (SOKT) is applied to remove range curvature. Then, a deramp function is proposed to reduce the order of coupling and remove the residual first-order coupling. In the following, Doppler parameters can be obtained through range IFFT and azimuth FFT. The proposed method is capable of accumulating the energy of ARU target's trajectory into a peak and obtaining Doppler parameters in range-Doppler domain with low computational complexity. Furthermore, it has better antinoise performance comparing with the Hough transform and Radon transform. The effectiveness of the proposed method is validated by simulated data for ground moving targets.

Airborne radar forward-looking imaging is of great significance in many remote sensing applications. However, the existing synthetic aperture radar (SAR) and Doppler beam sharpening (DBS) imaging techniques are incapable of forward-looking imaging. The real aperture radar (RAR) using a scanning antenna can provide forward-looking images, but suffers from coarse azimuth resolution. In this paper, we extend the iterative adaptive approach (IAA) to forward-looking super-resolution imaging. Different from the conventional forward-looking convolution model, both the Doppler phase and antenna convolution are considered in the new model, allowing for more accurate reconstruction of the forward-looking imaging scenario when applying the IAA. Simulation results demonstrate that the IAA-based super-resolution imaging can overcome the deficiencies of the SAR and DBS techniques in forward-looking imaging direction.

Synthetic aperture radar and its application in remote sensing have been an international focus in recent years. Feature extraction is a key step in SAR automatic target recognition. In this paper, 2DLSDDP is proposed for dimensional reduction and feature extraction, which is based on manifold learning theory. 2DLSDDP preserves not only local and class information of the original dataset, but also intrinsic geometric information of the image. In addition, it seeks a proper clustering direction in the neighborhood for each sample in feature extraction, which enhances the discriminative capability of the method. Compared to other SAR imagery feature extraction methods, the experiments based on MSTAR database show that the proposed method improves the recognition performance.

The accuracy of Doppler centroid estimation affects the target location and the scene mosaic in the Doppler beam sharpening imaging. Though it can be measured by different sensor instruments of servo, attitude, inertial, the inaccurate measurements decrease the imaging performance. In this paper, a Doppler centroid estimation method based on the morphological edge detection is proposed to obtain the Doppler centroid from the received data with loose-measured parameters. The Doppler frequency in the forward-looking region is symmetrical, but the target carries the highest Doppler frequency. The characteristic can be vividly reflected in range Doppler domain with morphological edge. The Doppler centroid can be estimated via the detection of the edge. The results of centroid estimation and Doppler beam sharpening imaging are given to verify the performance of the proposed method.

Raw data generation of SAR echo is important for the evaluation of both imaging algorithms and SAR system design. Common time-domain echo generation methods (represented by projection method) are accurate for most of SAR modes but usually inefficient, while the existing FFT-based methods can only adapt to some relatively ideal configurations for bistatic SAR. Hence, this article will focus on FFT-based methods and try to solve the limitation of adaptability. For bistatic SAR with complicated configuration or maneuvering platform trajectories, 2-D spatial variation and difficulty on obtaining the 2-D spectrum of echo signal are the major problems to be faced with. To solve these problems, this article proposes an efficient raw data generation method based on 2-D inverse wavenumber mapping. This method first performs a 2-D FFT on the raw image to generate its 2-D wavenumber spectrum, then induces the space-variant phase using the 2-D inverse wavenumber mapping. Simulation results are presented to verify the validity of the proposed method.

Identifying the aspect for a given target is an important issue in synthetic aperture radar (SAR) image interpretation. A new SAR target aspect identification method based on machine learning theory is proposed in this paper. First, the aspect angles of the SAR target are discretized, and the spatial relationships of the neighborhoods of the SAR target samples are established. Then an optimal linear mapping is solved based on the proposed subspace aspect discriminant analysis. The samples will be projected into a low-dimensional space and be of a better aspect identifiability than in their original space. Finally, the projected samples are fed into a multilayer neural network, and the aspects of the SAR targets will be indicated. Experimental results have shown the superiority of the proposed method based on the moving and stationary target acquisition and recognition (MSTAR) data set.

As for the bistatic forward-looking synthetic aperture radar (BiForSAR) imaging, frequency-domain imaging algorithms integrated with autofocus is advisable to get a well focused BiForSAR image in the presence of uncompensated motion errors. Nevertheless, a severe drawback of the autofocus algorithms is that they are only capable of removing one-dimensional azimuth phase errors. In range direction, there exists nonsystematic range cell migration (NsRCM) originated from motion errors and RCM correction procedure in frequency domain imaging algorithms as well. In this paper, the causes of different NsRCM in BiForSAR imaging are analyzed, and the corresponding relationship between different NsRCM is deduced. Based on the analyses, we propose an autofocus NsRCM correction scheme for BiForSAR imagery using frequency domain imaging algorithms, being capable of eliminating the range-dependent NsRCM.

This paper proposes a space variant-based maximum a posteriori (MAP) angular super-resolution algorithm for highspeed moving real-beam scanning radar. Firstly, the aberration model of the antenna modulation function is established through analyzing the relationship between the scanning angle and the sight angle. Afterwards, an efficient piecewise constant model is formulated for the sake of reducing the computational complexity and restoring cost. Finally, the space variant-based MAP algorithm is derived based on the aberrant model. Simulation experiments demonstrate that the proposed method can improve the super-resolution performance of the high-speed platforms more efficiently than the traditional MAP method.

Despeckle is a key post-processing step for Synthetic Aperture Radar (SAR) images [1], because speckles severely affect visual quality and intelligent interpretation. However the existing despeckle methods have a problem in texture preservation when the speckles are reduced. To solve this problem, image heterogeneity measurement is utilized to improvement the non-local means (NLM) despeckle method in this paper. First, an anisotropic window is constructed using coefficient of variation (CV), which is used for heterogeneity measurement. The anisotropic window are used to replace the isotropic window for the similarity measurement in NLM method. Furthermore, for weight function, an adaptive filter parameter based on heterogeneity measurement is designed to adjust the despeckling degree for different regions. The simulation results show that the proposed method outperforms than the existing algorithms.

For high-quality synthetic aperture radar (SAR) processing, Doppler centroid estimation is an essential procedure that can help to solve the SAR platforms motion parameters conversely, construct azimuth matching function and is important for target location as well. In this paper, a Doppler centroid estimator based on phase center tracking is proposed for SAR. An evaluation function is designed to measure the estimated Dopper centroid accuracy. In order to improve the calculation accuracy and computational efficiency, the evaluation function is solved by binary stage optimization to estimate the Doppler centroid. Simulations validate the effectiveness of this method.

When the target has large size or the target moves irregularly, traditional inverse synthetic aperture radar (ISAR) imaging method will lead to a poor image quality due to space variate and migration of echo envelop. In this paper, a novel method based on Back Projection (BP) and Particle Swarm Optimization (PSO) is proposed, which can achieve high quality images under the circumstances of irregular motion, large target and low signal to noise ratio (SNR). First, Target motion is modeled as a turntable, then the translational motion and rotational motion are modeled as two polynomials. Entropy of coherent superposition value of part of the imaging scene pixels based on BP algorithm is utilized as the evaluation function to estimate the polynomial coefficients based on an optimization algorithm such as PSO. Once the polynomial coefficients are estimated, a high quality image of the whole scene can be obtained by BP algorithm. The simulation results verify the effectiveness of the proposed method.

Due to finite of information dimension of single bistatic synthetic aperture radar (SAR), to expand more information about ground objects we research multistatic SAR which can be decomposed into groups of bistatic SAR. It is known that scattering properties of the different viewing angles is different. In this paper, we focus on system of single transmitter and triple receivers and use polar format algorithm(PFA) to obtain images of ground objects for the triple receivers. A geometric distortion correction method is proposed due to elevation of ground objects. After the distortion correction, the three SAR images are registrated, then we put images into red, green, blue(RGB) channels respectively to realize fake color synthesis, and thus realize information fusion.

The conventional deconvolution approaches which just rely on amplitude information behave worse when the radar platform speed is fast, and the approaches which just use the doppler phase caused by the platform moving can't achieve forward-looking imaging. To achieve forward-looking angular super-resolution for moving radar platform, in this paper, a complex deconvolution method which utilizes both amplitude and doppler phase information is proposed. The complex convolution matrix is constructed through the corresponding relation between the amplitude and doppler phase. The truncated singular value decomposition (TSVD) method is applied to suppress noise amplification to achieve deconvolution. Simulations demonstrate that the proposed method can achieve forward-looking angular super-resolution for moving radar platform.

It is great significant to achieve clear outline reconstruction for radar forward-looking imaging. In this paper, we apply the total variation (TV) function as the regularization term operator to obtain the forward-looking imaging with clear outline. Firstly, we establish the deconvolution model, by which the forward-looking super-resolution imaging problem is converted into inverse problem. Then, taking the TV function as regularization constraint term, we construct the objective function to solve the inverse problem. Finally, we obtain the minimum of the objective function, by which we can achieve radar forward-looking super-resolution imaging with clear outline. Simulations verify effectiveness of the proposed method in reconstructing the outline of targets.

This paper presents an angular superresolution method based on maximum a posterior (MAP) criterion to improve the azimuthal resolution of forward-looking scanning radar in the background of sea clutter. Firstly, in consideration of the statistical property of sea clutter, the Rayleigh distribution is employed to express the likelihood function. And then, the generalized Gaussian constraint is used as prior information about targets for better property of noise suppression and positional accuracy. Finally, the iterative expression is derived to recover the scattering coefficient of original sea-surface targets. Compared to the conventional Bayesian approaches, The results of simulation experiment are given to verify the superior performance of proposed method.

Oil spill will cause severe ecological disasters and enormous marine environment damages. we consider a robust and fast oil spill candidate detection problem for oil spill recognition systems for synthetic aperture radar (SAR) imagery. In this paper, we propose a automatic detection method called thresholding-guided maximally stable extremal regions (T-GMSERs) algorithm. First, thresholding approach is utlized to learn model parameters automatically. Candidate regions are extracted by using the maximally stable extremal region (MSER) detector. Then, we label each candidate region to obtain a binary potential target pixel map. Finally, the detection results are acquired by maximal stable criteria from the corresponding region map. Simulation based on satellite-borne data illustrates that the proposed algorithm obtains more precise detection performance without increasing the computational complexity.

This paper presents an adjustable angular superresolution method to realize high azimuthal resolution of forward-looking area in scanning radar imaging. Firstly, the received signal in azimuth dimension is established as the convolution model of target scattering coefficient and antenna pattern. Then based on the Poisson statistic assumption, the Generalized Gaussian distribution as prior constraint is used in the maximum a posterior (MAP) method due to the adjustability of dispersion parameter. At last, how to choose suitable dispersion parameter is discussed for better superresolution performance of different scenes. The simulations and experimental results are given to verify the performance of proposed method.

Spatial resolution plays a key role in evaluating the quality of synthetic aperture radar (SAR) images. Due to its flexible geometry and configuration, bistatic SAR (BSAR) shows much more complex resolution characteristics than mono static SAR, such as coexistence of non-square and nonorthogonality. Traditional comprehensive metrics of spatial resolution ignored the non-square property. SAR-based template matching is widely used in various fields such as scene-matching-aided navigation and its performance is influenced by the spatial resolution of the SAR image. In this paper, the non-square property of the resolution of BSAR is analysed by the resolution ellipse. Then the template matching method and the metric of performance are presented. Finally, template matching performance of BSAR systems with resolutions of different non-square degree but same resolution cell area is studied and simulated.

Bistatic forward-looking synthetic aperture radar (BFSAR) is a promising technique for applications such as automatic navigation and landing in all weather conditions and in day and night. Conventional echo models are no longer valid for the BFSAR mounted on the emerging hypersonic vehicle (HSV) due to its high-speed large-acceleration motion characteristics. In this paper, by taking the nonlinear maneuvers of the HSV receiver during pulse propagation into account, a closed-form non-stop-and-go echo model named maneuvering-during-pulse-propagation echo model is proposed for HSV-BFSAR. The model errors are analysed theoretically. The simulation verifies the validity of the proposed model.

In this paper, the Cramér-Rao Bound (CRB) for scanning radar sensing is investigated, providing an algorithm-independent bound on the angular estimation error. Based on the deterministic signal model, we first derive a numerical CRB for unknown real signal parameters. Then, the approximate closed-form expression of CRBs are further provided for the single target case. Meanwhile, the potential estimation error of various classical super-resolution sensing methods are quantitatively investigated in this paper, and compared with the presented CRB.

In forward-looking scanning radar imaging, the azimuth resolution can be improved by adding the sparse constraint. However, the azimuth resolution is limited with noise influence by traditional sparse regularization methods. In this paper, we propose a Bayesian super-resolution method that solves the L1regularization problem using the split Bregman algorithm. This method decouples L1and L2norms for the independence of them to reduce the computational complexity. The simulations verify that the proposed algorithm provides a better resolution and de-noising ability compare with conventional methods.

Geosynchronous (GEO) spaceborne-airborne multistatic synthetic aperture radar (GEO MulSAR) is more flexible and accessible in remote sensing applications because of the high-altitude illuminator and the separation of the receivers and transmitter. In addition, the information obtained by the multiple airborne receivers can be fused to enhance the spatial resolution. However, the fused spatial resolution severely depends on the applied multistatic topology. To achieve the optimal fused spatial resolution by properly adjusting the imaging topology, a topology design method is proposed in this letter. First, the spatial resolution model of GEO MulSAR is given, and the dependence of the spatial resolution on the multistatic topology is analyzed in detail. Then, a topology design method is proposed to obtain the best multistatic topology that simultaneously optimizes the resolution cell area and resolution disequilibrium factor. Finally, the simulation results validate the effectiveness of the proposed method, and some insights into designing the multistatic topology are given.

Deconvolution techniques provide efficient implementations for super-resolution imaging for forward-looking scanning radar. However, deconvolution is normally an ill-posed problem, and the solution is extremely sensitive to noise. From a statistical perspective, maximum likelihood (ML) methods are able to condition the ill-posed problem into a well-posed one. Nevertheless, traditional ML methods only consider the amplitude of the echo and by ignoring the phase that do not adequately model the radar imaging system. In this letter, an I/Q-channel modeling ML method is proposed for forward-looking scanning radar. First, the probability model of the echo is deduced by jointly considering noise in the I and Q channels. Then, a probability density function of the received data is deduced and used to formulate the likelihood function. Finally, the targets can be precisely estimated by maximizing this likelihood function. The results of simulations and experiments are provided to illustrate the effectiveness of the proposed method.

In synthetic aperture radar (SAR) imagery, images of ground moving targets (GMTs) are smeared, distorted, and shifted. Current GMT imaging methods are mostly based on range-Doppler algorithms, which have two main drawbacks: 1) coupling between range cell migration correction (RCMC) and Doppler parameter estimation and 2) cross terms degrade the performance of the nonlinear estimation methods. In this paper, an optimal 2-D spectrum matching method for SAR GMT imaging is proposed. The main innovation or advantage of this method is that the GMT imaging problem is transformed into a constrained optimization problem, and differential evolution is applied to guarantee a high-processing efficiency. As they are associated with the Doppler centroid variation compensation and range shift compensation processing, all GMT point scatterers can be well focused and well located. Compared with the current methods, the improvements exhibited by this method include three main benefits: 1) RCMC and Doppler parameter estimation can be simultaneously accomplished; 2) both along- and cross-track GMT velocities can be simultaneously estimated; and 3) this method can be applied to both monostatic and bistatic SARs. Numerical simulations and experimental data processing have verified the effectiveness and robustness of the proposed method.

It is a feasible and promising way to utilize deep neural networks to learn and extract valuable features from synthetic aperture radar (SAR) images for SAR automatic target recognition (ATR). However, it is too difficult to effectively train the deep neural networks with limited raw SAR images. In this paper, we propose a new approach to do SAR ATR, in which a multiview deep learning framework was employed. Based on the multiview SAR ATR pattern, we first present a flexible mean to generate adequate multiview SAR data, which can guarantee a large amount of inputs for network training without needing many raw SAR images. Then, a unique deep convolutional neural network containing a parallel network topology with multiple inputs is adopted. The features of input SAR images from different views will be learned by the proposed network layer by layer; meanwhile, the learned features from the distinct views are fused in different layers progressively. Therefore, the proposed framework is able to achieve a superior recognition performance, and requires only a small number of raw SAR images for network training samples generation. Experimental results have shown the superiority of the proposed framework based on the Moving and Stationary Target Acquisition and Recognition data set.

Multiview synthetic aperture radar (SAR) images could provide much richer information for automatic target recognition (ATR) than from a single-view image. It is desirable to find optimal SAR platform flight paths and acquire a sequence of SAR images from appropriate views, so that multiview SAR ATR can be carried out accurately and efficiently. In this paper, a novel optimization framework for multiview SAR ATR is proposed and implemented. The geometry of the multiview SAR ATR is modeled according to the recognition mission and flight environment. Then, the multiview SAR ATR is abstracted and transformed into a constrained multiobjective optimization problem with objective functions considering the tradeoffs between recognition performance and efficiency and security. A specific approach based on convolutional neural network ensemble and constrained nondominated sorting genetic algorithm II is employed to solve the multiobjective optimization, and optimal flight paths and corresponding imaging viewpoints are obtained. The SAR sensor can thus choose an applicable flight path to acquire the multiview SAR images from different tradeoff solutions according to application requirements. Finally, accurate recognition results can be obtained based on those multiview SAR images. Extensive experiments have shown the validity and superiority of the proposed optimization framework of multiview SAR ATR.

In general, autofocus methods integrated with frequency-domain imaging algorithms are instrumental to obtain a well-focused bistatic forward-looking synthetic aperture radar (BFSAR) image in the presence of motion errors. Nevertheless, before applying autofocus methods to correct the azimuth phase errors, range cell migration (RCM) should be eliminated by the RCM correction (RCMC) procedure in frequency-domain imaging algorithms. With motion errors being taken into account, there always exists some residual nonsystematic RCM (NsRCM), which refers to the residual migration components after the RCMC procedure. For the conventional side-looking SAR, NsRCM is caused by motion errors. On the other hand, NsRCM of BFSAR is originated from motion errors before RCMC and the NsRCM amplified by the RCMC procedure. In this paper, we analyze the different types of NsRCM in BFSAR imaging and their relationship. Based on the analyses, we propose an autofocus NsRCM correction scheme for BFSAR imagery using frequency-domain imaging algorithms that can eliminate the range-dependent NsRCM. The proposed scheme consists of three steps. First, for the BFSAR data after range compression and RCMC, a division procedure is carried out in the azimuth direction. The subblocks with the highest signal-to-clutter ratio along the range direction are selected after the azimuth segmentation procedure. Second, for the selected subblocks, the total NsRCM is estimated based on the minimum-entropy criterion. Based on the estimation results, different parts of the NsRCM are obtained by solving an ordinary differential equation. Third, a two-step compensation of the NsRCM is executed to reach the spatially variant correction. Simulations and experimental results are provided to demonstrate that our proposed scheme is effective for BFSAR imaging.

Recently, the generalized sparse iterative covariance-based estimation algorithm was extended to allow for varying norm constraints in scanning radar applications. In this paper, further to this development, we introduce a wideband dictionary framework which can provide a computationally efficient estimation of sparse signals. The technique is formed by initially introducing a coarse grid dictionary constructed from integrating elements, spanning bands of the considered parameter space. After forming estimates of the initially activated bands, these are retained and refined, whereas nonactivated bands are discarded from the further optimization, resulting in a smaller and zoomed dictionary with a finer grid. Implementing this scheme allows for reliable sparse signal reconstruction, at a much lower computational cost as compared to directly forming a larger dictionary spanning the whole parameter space. Simulation and real data processing results demonstrate that the proposed wideband estimator offers significant computational savings, without noticeable loss of performance.

High-resolution scanning radar mapping of the surface is an effective tool for addressing concerns in local environmental and social investigation fields. Regrettably, the azimuth resolution of a scanning radar is constrained by the antenna beamwidth. Multiple super-resolution approaches have been applied to the scanning radar to enhance the azimuth resolution, but they suffer from limited resolution improvement. In this paper, a methodology to derive surface estimates from the scanning radar at an improved azimuth resolution is proposed. We first consider the truncated spectrum by discarding the unreliable frequencies to suppress the noise amplification. Then, based on the iterative adaptive approach (IAA), a novel inverse filtering method is formulated to obtain lower sidelobes and a higher resolution. Finally, by taking advantage of the Fourier property of the steering matrix and the Toeplitz structure of the covariance matrix, we exploit the Gohberg-Semencul representation and the data-dependent trigonometric polynomials to derive a fast IAA (FIAA)-based inverse filtering to mitigate the computational burden. Simulation results and real data processing demonstrate that the proposed FIAA-based inverse filtering outperforms the existing super-resolution approaches in resolution improvement and results in a higher computational efficiency.

In this paper, under the situation of multiple interference regions, an optimal antenna placement problem for a distributed Multi-Input Multi-Output (MIMO) radar is studied. Considering multiple interference regions, we solve the antenna placement problem by utilizing antenna placement method based on Multi-Objective Particle Swarm Optimization (MOPSO). However, it is not clear when to stop the iteration for which no knowledge about the optimum result is available. Hence, computational resource may be wasted over iterations. Nevertheless, time and computational resource is limited in real application. Therefore, to obtain the optimal placement result with limited time and computational resource, an iteration convergence criterion based on interval distance is proposed. The iteration convergence criterion can be used to stop the optimization process efficiently when the optimal antenna placement algorithm reaches the desired convergence level. Finally, numerical results are provided to verify the validity of the proposed algorithm.

This paper describes a new technique for generating bistatic sea clutter returns based on the compound K-distribution model for clutter amplitude statistics. The technique adopts the computational electromagnetic (CEM) method to calculate bistatic sea clutter reflectivity by the given bistatic geometrical relationship, aiming at obtaining the parameters of the distribution. Then the theory of spherically invariant random processes (SIRP) is used to generate the returns of the bistatic sea clutter following compound K-distribution. This study can be used to evaluate the bistatic radar signal model and predict system detection performance in the sea clutter environment. Simulation results verify the proposed technique.

This article presents an adaptive large-size moving-target imaging technique for bistatic forward-looking SAR (BFL-SAR). The main problems of this issue are that not only the echo characteristics (including range cell migration and Doppler parameters) are unknown, but also the spatial-variances of these characteristics are nonlinear for different point-scatterers. The proposed technique relies on a proper processing of the data aiming at, first, to correct the range walk by applying keystone transform over the whole received echo, and then, the relationships between the unknown high-order RCM, the nonlinear spatial-variances of the Doppler parameters, and the speed of the mover, are established. After that, using an adaptive NLCS technique, not only the unknown high-order RCM can be accurately corrected, but also the nonlinear spatial-variances of the Doppler parameters can be balanced. Numerical simulations show the effectiveness of this adaptive large-size moving-target imaging technique to be employed in BFL-SAR frameworks.

Synthetic aperture radar (SAR) has been widely used in remote sensing. Feature extraction is a crucial step in SAR automatic target recognition (ATR). In this paper, Kernel Marginal Sample Discriminant Embedding (KMSDE) is proposed, which is based on kernel trick and manifold learning theory. In feature extraction via KMSDE, the original dataset is mapped to high dimensional space and manifold learning theory is introduced for dimensional reduction. KMSDE preserves local and class information of the original dataset, as well as gathers the with-class samples and separates between-class samples in the low-dimensional space. In addition, it employs information related to each sample's location in the dataset, which enhances the discriminative capability of the method. Compared to other SAR imagery feature extraction methods, the experiments based on MSTAR database show that the proposed method improves the recognition performance.

Recognizing the category attributes from the real world targets is one of the most challenging and attractive fields in synthetic aperture radar (SAR) application. It is an important issue to explore the spatial distribution characteristics of multi-aspect angle imagery in synthetic aperture radar automatic target recognition (SAR ATR). In this paper, we will research the spatial structure of multi-aspect angle SAR imagery through a visualization approach with real SAR data. Based on nonlinear dimensionality reduction, the representation of SAR samples is revealed in the low-dimensional Euclidean space, and the the nonlinear manifold distribution of multi-aspect angle SAR imagery is discovered. Besides, the regularity of that spatial distribution is summarized, i.e. the intrinsic structure of SAR images is parameterized by the aspect angles. The results of our research can provide a theoretical basis for SAR image classification and recognition algorithm designing.

Regularization technology can be utilized to improve the azimuth resolution for forward-looking scanning radar (FLSR). Among various regularization methods, L1 norm constrained method is usually adopted for its strong ability in resolving the sparse targets. Nevertheless, the solution of L1-norm constrained regularization method (L1-CRM) is sensitive to noise and the iterations would quickly diverge from the desired result if too many iterations are performed. In this paper, a regularization imaging method via joint L1-L2 norm constraint is proposed. By combing the L1 norm constraint and L2 norm constraint together, a new objective function is obtained and then the popular fast iterative threshold/shrinkage (FIST) method is adopted to minimize this nonsmooth convex function. This new method can not only have strong ability in resolving the sparse targets, but also be more robust to the noise. Simulations are carried out to demonstrate the effectiveness of the proposed method in terms of resolving ability and noise robustness.

Angular super-resolution imaging plays a significant role in the area of the scanning radar imaging. Some deconvolution methods are used to realize the angular super-resolution on scanning radar. However, the ill-posed nature of the deconvolution problem means difficulties and inaccuracies in the search for the solution. In this paper, we present a novel method for angular super-resolution imaging in scanning radar using the alternating direction method for solving the constrained optimization problem. To this end, we first formulate the angular super-resolution problem as deconvolution task and then convert it to a constrained optimization problem by incorporating the prior information of the targets. We then attack the constrained optimization problem in augmented Lagrangian framework using an alternating direction method, leading to the algorithm that can be implemented easily. It is shown in a serious simulation that the proposed algorithm outperforms a number of existing deconvolution algorithms in terms of stability and precision.

High cross-range resolution of forward-looking region is the key problem of radar imaging. This paper presents a multi-beam system to extend high resolution imaging domain based on the digital beamforming (DBF) technique. The Doppler bandwidth is increased in forward-looking direction while the symmetrical Doppler domain could be significant reduced. Then matched filter technique is employed to deal with the composed received echo. The feasibility and effectivity of this strategy were verified by simulation results.

Breaking through the forward-looking limitation of monostatic synthetic aperture radar (SAR), bistatic forward-looking SAR (BFL-SAR) brings much potential in areas such as automatic navigation and landing in bad weather conditions. For many cases of BFL-SAR mounted on maneuvering-platform (MBFL-SAR), neither the flight path is straight nor the topography is flat. Conventional imaging algorithms designed for linear trajectory and flat topography fail to handle this situation. In this paper, a full-aperture efficient extended nonlinear chirp scaling algorithm with topography compensation is proposed for the data processing of MBFL-SAR. Simulation is presented to verify the validity of the proposed algorithm.

Target recognition based on forward-looking imaging has many potential applications. However, the conventional algorithms fail to locate targets accurately due to the low resolution of forward-looking radar images. Meanwhile, the conventional algorithms always suffer from high computational complexity and cannot satisfy the requirement of real-time processing for high-speed platform. This paper proposes a novel target recognition method based on morphological and spatial features. The algorithm comprises of initial matching and dual verification algorithms based on image gray scale and a priori position information. It is demonstrated that the proposed algorithm can work well for the forward-looking radar images with coarse resolution and enjoy higher computational efficiency. Simulation and real data processing validates the superior performance of the proposed algorithm.

In bistatic forward-looking synthetic aperture radar (BFSAR), conventional autofocus algorithms cannot estimate the phase errors accurately when the range walk is compensated in the azimuthal time domain. This problem stems from the influence of the azimuth-variant Doppler coefficients after linear range cell migration correction in azimuthal time domain. In principle, nonlinear chirp scaling processing can be carried out to remove the azimuth variance. Nevertheless, Doppler azimuth variations of the Doppler parameters induced by the motion errors are not taken into consideration and serious defocus would emerge in the final image. To cope with such a problem, an estimation-evaluation-equalization scheme is proposed before conventional autofocus for BFSAR. Different from the conventional autofocus method, the azimuth-variant Doppler coefficients are additionally estimated and equalized before autofocus, and as a consequence, phase errors can be precisely estimated. The BFSAR data processing results demonstrate the validity of the proposed method on the improvement of autofocus in BFSAR.

Recently, the iterative adaptive approach (IAA) was adopted to allow for the estimation of high-resolution scanning radar images. In this letter, we further develop this approach by introducing a range-recursive IAA (IAA-RR) formulation allowing for a computationally efficient updating of the resulting estimates along range. Besides exploiting the rich matrix structure to mitigate the computational complexity for each iteration, the correlation between adjacent range cells is exploited to accelerate the convergence of the IAA iterations. When an additional range measurement becomes available, further acceleration is available by exploiting the estimates already formed for the adjacent range cells. Compared with the existing fast IAA implementation, the proposed IAA-RR is shown to offer significant computational savings, without noticeable loss in performance. Numerical results illustrate the superior performance of the proposed IAA-RR algorithm.

This paper considers robust vehicle logo recognition (without aiming at accurate location) for intelligent transportation systems. We propose a recognition-before-location framework for multi-scale vehicle logos which exploits a directional SIFT flow parsing method. We extract dense SIFT descriptors of different standard vehicle logos. An improved matching method is proposed to obtain a directional SIFT flow from standard logo models for vehicle images. Our vehicle logo recognition algorithm is based on dense SIFT matching energy and SIFT flow consistency. We verify the accuracy of vehicle logo recognition and the robustness for multi-scale logo images on various real data.

Target image database is of great significance in SAR automatic target recognition (ATR). Recently, some convenient and low cost approaches of database simulation were proposed. However, the similarity between virtual SAR images obtained by these simulation approaches and real SAR images is still under study. To solve this problem, we will model the virtual target with three-dimensional (3D) modeling methods, and acquire SAR image via the simulated RCS data which is generated by computational electromagnetic software. Then, we propose a method to measure the similarity between the virtual and real SAR images, which provided better support for data training and recognition of virtual target. Experiment results demonstrate the formation of the virtual SAR images and validate the effectiveness of our proposed method.

In this paper, SAR MT imaging and velocity estimation method is proposed. The validity of this method is verified by numerical simulations. The main idea behind this method is to transform the PE problem to be a SOP problem. The advantages of this method include two main aspects: (i) This method can handle the MT imaging problem for different SAR modes, such as mono-static SAR, bistatic SAR, etc.; (ii) Both the along-track and cross-track velocities of theMT can be simultaneously estimated. In addition, since the focusing processing is conducted in 2D spectrum domain and the optimal criterion is the local minimum entropy, this method don't need to find a dominated point scatterer during the process.

Deconvolution techniques can be utilized to realize angular resolution enhancement for real-beam scanning radar (RBSR). However, most deconvolution algorithms are sensitive to noise and time consuming. In this paper, an accelerated iterative shrinkage/thresholding (AIST) algorithm is proposed to overcome these disadvantages. AIST is developed from iterative shrinkage/thresholding (IST) algorithm which can reduce the noise sensitivity through shrinkage/thresholding operation. And a vector extrapolation method is adopted to accelerate the IST algorithm. This acceleration method makes a prediction before IST iteration step and thus can converge faster than IST. Simulations are carried out to demonstrate the AIST algorithm can effectively accelerate IST algorithm without performance loss.

In this letter, an improved regularized iterative adaptive approach (IAA) is proposed for scanning radar angular superresolution. Because the IAA requires matrix inversion, the increasing condition number of the covariance matrix leads to the ill-posed problem of the IAA. Based on this reality, the diagonal loading method is introduced to solve the ill-posed problem. Because the loading value controls the tradeoff between the azimuth resolution and noise amplification, we use the radiometer uncertainty principle to determine the optimum loading value. When compared with the existing angular superresolution approaches, the proposed regularized IAA is shown to provide significant resolution improvement. Numerical results illustrate the superior performance of the proposed regularized IAA.

Bistatic forward-looking synthetic aperture radar (BFL-SAR) is a kind of bistatic SAR system that can image forward-looking terrain in the flight direction of an aircraft. Until now, BFL-SAR imaging theories and methods have been researched for stationary targets. Unlike the stationary target, the motion of a ground-moving target (GMT) induces unknown range cell migration and additional modulation of the azimuth signal. Thus, to finely image the GMT, one must obtain its velocity parameters accurately, but they are usually unknown. In this paper, a novel GMT imaging and velocity estimation method, which is based on mismatched compression, is proposed for BFL-SAR without a priori knowledge of the GMT's velocity parameters. The main idea behind mismatched compression is to use a presumed azimuth reference function for performing correlated operation with the azimuth signal of the GMT. In general, the Doppler parameters of the presumed azimuth reference function are different from those of the GMT's azimuth signal because the velocity parameters of the GMT are unknown. Therefore, the correlation operation referred to earlier is actually mismatched compression, and the resulting image is shifted and defocused. The shifted and defocused image is utilized to get the real Doppler and velocity parameters of the GMT. The advantage of this method is that not only the GMT can be well focused but also the GMT's velocity can be simultaneously obtained. In addition, this method needs only monochannel antenna. The proposed BFL-SAR GMT imaging and velocity estimation method is validated by numerical simulations.

With appropriate geometry configurations, bistatic synthetic aperture radar (SAR) can break through the limitations of monostatic SAR on forward-looking imaging. Owing to such a capability, bistatic forward-looking SAR (BFSAR) has extensive potential applications. In BFSAR, the compensation of the spatially variant motion errors is of great significance to get a well-focused image. In this paper, first, the spatial-variance properties of motion errors are analyzed analytically and quantitatively. Different from the side-looking monostatic and bistatic SAR, 2-D space-variant motion errors should be taken into consideration in BFSAR. The 2-D spatial variance of the motion errors can be categorized into two parts, range-variant motion errors of the transmitter and azimuth-variant motion errors of the receiver. Moreover, these two parts are independent of each other. Based on this property analysis, second, a motion compensation (MoCo) approach with cubic-order processing is proposed to deal with the spatially variant motion errors in BFSAR. In the cubic-order processing, the first-order MoCo is performed to correct the spatially independent motion errors on the raw data. The second-order MoCo is accomplished on the non-range-cell-migration (RCM) data to deal with the range-variant errors. After the second-order MoCo, since the signal direction of the non-RCM data coincides with the variant direction of the uncompensated phase errors, the azimuth-variant motion errors and slow time signal are coupled together. To cope with such a problem, the slow time signal is transformed into the direction perpendicular to the azimuth by a novel procedure named azimuth-slow time decoupling. At this stage, the coupling between the azimuth-variant motion errors and slow time signal has been eliminated. Azimuth-variant motion errors can be corrected precisely. Simulation and experimental results verify the effectiveness of the proposed method.

Geosynchronous synthetic aperture radar (GEO-SAR) offers new opportunities for continuous Earth observation missions with large coverage and short revisit cycle. The unique features of GEO-SAR present huge potentials for bistatic observation applications. In this paper, the concept and advantages of GEO bistatic SAR (GEO-BiSAR) are first investigated. The system consists of a GEO illuminator and an airborne receiver, such as an airplane or a near-space vehicle. Compared with a monostatic GEO-SAR system, the bistatic configuration can provide finer spatial resolution and higher signal-to-noise ration (SNR) with less system complexity. The spatial resolution characteristics are then analyzed based on generalized ambiguity function, where the time-varying GEO velocity, Earth rotation, and ellipsoid Earth surface are taken into consideration. Meanwhile, the bistatic SNR is analyzed using the integration equation model. In this paper, the mission design for GEO-BiSAR aims at identifying a set of receiver flight parameters and bistatic configurations to obtain the desired spatial resolution and SNR. Based on the desired imaging performance of a specific application background, the mission design process is modeled as a nonlinear equation system (NES). Finally, a mission design method based on fast nondominated sorting genetic algorithm is proposed to solve the NES and obtain multiple optimal solutions to guide the receiver flight missions. Examples of the mission design process are given to validate the effectiveness of the proposed method. The results of the mission design can be conveniently used to guide the receiver flight mission for the desired imaging performance, which is highly desirable in practical applications.

With the geosynchronous synthetic aperture radar (SAR) satellite as the transmitter, the unmanned aerial vehicle (UAV) can passively receive the echo within the illuminated ground area and achieve 2-D imaging of the interested target. This SAR system, known as GEO-UAV bistatic SAR, is capable of autonomously accomplishing the bistatic SAR mission in rough terrain environments by prespecifying a path for the UAV receiver. In this paper, the GEO-UAV bistatic SAR system is first investigated. The practical advantages and spatial resolution are then analyzed in detail. The spatial resolution of GEO-UAV bistatic SAR is dependent on the observation geometry, which is determined by the UAV path. Therefore, the path planning for GEO-UAV bistatic SAR aims at identifying a set of optimal paths for the UAV receiver to travel through a 3-D terrain environment that simultaneously guarantees the safety of the UAV and achieves SAR imaging with optimized performance during the flight. The path planning is modeled as a constrained multiobjective optimization problem (MOP), which accurately represents the two main aspects for the path planning problem, i.e., UAV navigation and bistatic SAR imaging. Then, a path planning method based on a constrained-adaptive-multiobjective-differential-evolution algorithm is proposed to solve the MOP and generate multiple feasible paths for the UAV receiver with different tradeoffs between navigation for UAV and bistatic SAR imaging performance. The GEO-UAV bistatic SAR mission designer can choose a path from the solution set according to the application requirements, which makes the method more pragmatic.

Angular super-resolution of scanning radar is an important problem in radar system. Some deconvolution methods are used to realize the angular super-resolution in scanning radar. However, the ill-posed nature of the deconvolution problem leads to the noise amplification in the angular super-resolution image. This phenomenon brings the difficulty in signal detection and tracking. In this paper, a deconvolution algorithm based on truncated singular value decomposition is proposed that achieves the angular super-resolution and noise suppression in scanning radar. To this end, we first convert the angular super-resolution task in scanning radar as an equivalent deconvolution problem. Then, the cause of noise amplification is analysed, which leads to the truncation singular value method for solving the deconvolution problem. Simulation results demonstrate that the proposed method is effective in achieving angular resolution with suppressing the noise amplification.

To obtain high resolution image for real beam scanning radar, angular superresolution is studied, and a two-step scheme is proposed in this paper. In the scheme, using the threshold determined by CFAR-OSTU, the real beam data is divided into two parts firstly, and then maximum likelihood deconvolution with different number of iterations is conducted on the two parts of data. At last, the superresolution image is obtained by combining the two parts of data.

In the imaging processing of translational variant bistatic forward-looking SAR (TV-BFSAR), range cell migration correction (RCMC) is an essential procedure. Based on the azimuth variance property of the range cell migration (RCM), the keystone transformation is used for RCMC in TV-BFSAR. Before the RCMC, the ambiguity correction of the Doppler frequency must be done. However, the motion error makes the ambiguity correction hard to be realized. An iterative RCMC based on Keystone transformation scheme is proposed in this paper to overcome the challenge motion error. By searching for the correct Doppler ambiguity number based on the minimum entropy, this scheme corrects the RCM in TV-BFSAR. Simulations validate the effectiveness of this method.

The ground moving target (GMT) model has been widely employed in modern coherent radar systems, such as the synthetic aperture radar (SAR) and the bistatic SAR (BiSAR). For the coherent radar systems, GMT imaging necessitates the compensation of the additional azimuth modulation without a priori knowledge of the GMT's motion parameters. That is to say, it is necessary to estimate the Doppler parameters of the GMT before the azimuth compression processing. For conventional estimation methods, such as the map drift (MD) method and the phase gradient auto-focus (PGA) method, a searching procedure is necessary and leads to an expensive computational cost. In this paper, a Doppler parameter estimation method based on mismatched compression is proposed. One advantage of this method is that it doesn't need the searching procedure. In addition, another advantage of this method is that both the Doppler centroid and the Doppler frequency rate of the GMT can be simultaneously estimated according to the relationships among the Doppler parameters, the positional offset and the boarding width of the mismatched imaging result. The theoretical analysis and numerical simulations validate that the proposed method works well with different signal to noise ratio.

The angular superresolution is of great significance for scanning radar in forward-looking imaging. There are many techniques documented in literature to enhance the angular resolution, of which deconvolution method and power spectral density(PSD)methods are favored and attain many interests. In this paper, we focus on analyzing the advantages and challenges of PSD methods in comparison with the deconvolution method. Firstly, three typical PSD estimation approaches are introduced, followed with the comparison with deconvo-lution method that summarizes the advantages and challenges of PSD methods in theory. Simulations are provided in terms of coherence and number of snapshots, which presents the performance of different PSD methods and Lucy-Richardson deconvolution method, better demonstrating the advantages and challenges of PSD methods.

Ship detection in sea clutter environment using scanning radar is of vital importance, but with challenges due to low angular resolution. To solve the problem, an angular superresolution algorithm for radar imaging based on Bayesian deconvolution theory is proposed. Firstly, the statistic characteristics of the sea clutter are modeled using compound K-distribution. Then the signal model of radar echo in sea environment is formulated as the convolution of the antenna pattern and the reflectivity of the original scene plus the reflectivity of sea clutter. The ship detection task in sea clutter environment using the deconvolution method is converted into an equivalent maximum a posteriori estimation problem, which is solved using the optimization method in this paper. Simulation results demonstrate the validity of the proposed method in terms of ship detection in sea clutter environment.

This paper presents a deconvolution algorithm based on the Maximum likelihood (ML) criterion to realize azimuth angular superresolution of sea-surface target in the background of sea clutter. Firstly, the received signal of real-beam image in azimuth dimension was modeled as the convolution of antenna pattern and target scattering. Then the ML objective function was built according to the assumption that the sea clutter obeys Rayleigh distribution. Finally, the iterative expression was derived to recover the original scattering of sea-surface target. Compared to Poisson-based ML deconvolution method, the assumption of clutter distribution is more reasonable for the practical background. Simulations are given to verify the effectiveness of the algorithm.

This paper presents a sparse deconvolution algorithm based on the Maximum a Posterior (MAP) criterion to improve the azimuth angular resolution of scanning radar. For high resolution of few targets in large imaging scene, the sparse property was considered as the prior information to combine with the classic Richardson-Lucy (R-L) iterative algorithm to obtain sparse solution. Besides, the sparsity constraints also help to suppress the amplification of false target and noise of R-L algorithm. Simulations are given to verify the effectiveness of our algorithm.

A scheme of fast standard Capon beamformer (SCB) is proposed for scanning radar angular superresolution aiming at the computation burden caused by large swath mapping in azimuth. First, using the similar block tridiagonal property between the covariance matrix and Schur complement of its submatrix, the fast matrix inverse works in an improved divide and conquer (D&C) approach by recursively breaking down the problem of fast inverse into the same sub-problems. Secondly, based on the circulant structure of steering matrix, the SCB estimate of every assumed grids are rewritten by linear convolution. The resulting algorithm is shown to reduce the necessary computational load with one power without noticeable loss of performance.

Most of ever proposed scanning radar angular superresolution algorithms are iterative, but the optimum iterations are difficult to determine. Since the performance of them is related to the signal to-noise ratio (SNR), an accurate SNR estimation would be of great significance to provide reference for assistance of adaptive iteration termination condition analysis. In this paper, a data-aided (DA) SNR estimation approach is developed. This scheme first utilizes the known antenna pattern and identity matrix to construct the steering matrix of signals and noise. Then we introduce the iterative adaptive approach (IAA) to estimate SNR. Simulation validates that this scheme, termed as IAA-SNR can present an adaptive and effective SNR estimation for scanning radar.

Highly squint synthetic aperture radar (SAR) data focusing is a more challenging and difficult task than the side-looking SAR due to the strong 2-D coupling of echo signal induced by the imaging mode. Although several algorithms have been proposed, they fail to take into consideration the spatial variance of linear range cell migration (RCM). Moreover, the RCM correction (RCMC) process in range frequency and azimuth time domain by phase multiplication brings the problem of azimuth variation of the Doppler centroid, which has a great influence on the azimuth focusing. In this letter, an algorithm based on keystone transform (KT) and azimuth extended nonlinear chirp scaling (ENLCS) is derived to deal with these problems. A new method for higher order RCMC is derived to remove the residual RCM after KT. Then, azimuth ENLCS is performed to equalize the azimuth-variant Doppler centroid and FM rate. The simulation results verify the effectiveness of this algorithm.

In this letter, we demonstrate an optimization problem of antenna placement of distributed multi-input multi- output (MIMO) radar. To evaluate the surveillance performance of the radar system, a coverage ratio is proposed as a criterion. Since the problem is of extremely huge computational complexity due to its complicated objective function and high dimensionality, we propose a solution that contains two parts: 1) a low- complexity method to simplify the objective function; 2) a placement algorithm based on particle swarm optimization (PSO) to deal with the challenge of high dimensionality. We also analyse the computational complexity of our solution. Simulation results verify the validity and advantage in computational complexity of our solution. Our contributions include a novel optimization placement model of distributed MIMO radar and a computational efficient solution to establish the optimal positions of antennas.

The combination of frequency-modulated continuous-wave (FMCW) technology and bistatic synthetic aperture radar (SAR) promises a small, light-weight, cost-effective and high-resolution remote sensing system. In this paper, an Omega-k imaging algorithm based on linearization theory to focus the raw data of translational invariant (TI) bistatic FM-CW SAR is proposed. This method utilizes a bistatic point target reference spectrum (BPTRS) of generalized Loffeld's bistatic formula (GLBF), where the motion during the long signal duration time is taken into account. The proposed BPTRS significantly simplify the TI bistatic FMCW SAR processing. Based on the spatial linearization of GLBF, the Stolt transformation is further derived. Finally, simulation experiments are carried out to verify the effectiveness of the proposed method.

Scanning radar imaging has significant and extensive applications, such as surveillance, autonomous landing of aircraft, navigation, guidance. However, scanning radar imaging system suffers troublesomely antenna-induced poor angular resolution. In this letter, a scheme of angular superresolution based on maximum entropy (ME) framework is presented. Firstly, radar received signal in azimuth is modeled as convolution of the antenna pattern and targets' scattering coefficient approximately. Then, principle of angular super-resolution algorithm is deduced. The algorithm can endure disturbance of noise more significantly than conventional techniques. Simulations validate that this algorithm can effectively enhance radar angular resolution. At last, real scanning radar imagery data has proved effectiveness of the proposed method by comparing with Wiener filter technique.

Radar image resolution is a controlling factor in the radar imaging application. In this paper, we propose an approach to radar angular super-resolution through sparse deconvolution, which is able to increase the resolution of radar image beyond the limitation of system parameters. It relies on the optimization approach that enables to incorporate the prior information about the system and the statistical characteristics of scene. We first formulate the radar angular super-resolution problem as a constrained optimization problem and then convert it to an equivalent unconstrained optimization task using augmented Lagrangian method. We then solve the unconstrained optimization problem in the convex optimization framework using iterative method. Numerical experiments with real data demonstrate that the validity of the proposed method.

Current literatures and reports about Bistatic forward-looking SAR (BFSAR) imaging theories are mostly concentrated on static scene. In this paper, a novel GMT detection and imaging method in Doppler-rate domain for BFSAR is presented. The steps of this method involve bulk-deramp filtering operation, Keystone transform (KT), extend azimuth nonlinear chirp scaling (EA-NLCS) processing and the last detection step based on product second-order ambiguity function (PSAF). After the above steps of the method, if the PSAF of a certain range bin has more than one sharp peak, we can argue that there is a GMT in the relative range bin. Then using the GMT's Doppler rate estimated by PSAF, the echo of GMT can be focused, thus the goals of GMT detection and imaging for BFSAR are achieved. Numerical simulations verify the effectiveness of the proposed method.

One-Stationary bistatic forward-looking SAR (OSBFSAR) is a SAR system that using a geostationary satellite or a near-space low-speed platform as the transmitter and using an airborne or a missile as the receiver. Current literatures and reports about OSBFSAR are mostly about imaging theory and stationary scene imaging methods. In this paper, we propose an OSBFSAR moving target detection and imaging method with a linear antenna array. This method is associated with the two-dimensional spatial variation imaging technology of OSBFSAR and the extended velocity-SAR technology. Using this method, not only the stationary clutter can be suppressed but also the moving targets can be detected and focused well. Some numerical experiments are given to verify the validity of the method shown in this paper.

Ground moving target detection (GMTD) for squint synthetic aperture radar (SAR) is a more challenging task than that of the broadside SAR. The severe range and azimuth coupling induced by the squint mode combined with the unknown motion parameters makes the problem of indicating the moving targets more involved. In this paper, a novel ground moving target detection method for squint SAR is proposed. Firstly, an extended Keystone Transform (KT) is performed to remove the range cell migrations (RCMs) of both the stationary and moving targets. Secondly, the range compressed data undergoes two extended azimuth nonlinear chirp scaling (EANLCS) process separately with different sets of parameters based on the stationary scene. Then, after azimuth de-ramp processing, two images are obtained where all the targets are well focused except for the moving targets. Finally, by subtracting the amplitudes of the two images, the stationary targets are eliminated and the moving targets are detected. This algorithm is suitable for large squint angle cases and is computationally efficient. Simulation results verify the effectiveness of the algorithm.

Multi-pass synthetic aperture radar coherent change detection (SAR CCD) is a coherent change detection technique using multiple images acquired along different flight tracks. The defining characteristic of multi-pass SAR CCD is the high detection probability under a certain false alarm probability. Changes are detected in areas of low coherence between SAR images. Unfortunately, low coherence can result from phenomenon other than changes such as the difference between passes. In order to reliably detect small changes of interest, false alarm rates should be driven down through means of acquiring and processing images interferometrically. In particular the image pair must be acquired with careful control of the multi-pass imaging geometries and collection time. In this paper, we propose some data collection strategies for high quality multi-pass SAR CCD, which include the space strategies and the time strategies. The space strategies include some constraint conditions of the imaging geometries such as baseline, terrain slope, grazing angle and squint angle. The time strategies include the choice of pass number and the time interval between different passes. The images used for multipass SAR CCD should be acquired in accordance with these strategies for better detection performance.

The butterfly algorithm is introduced to achieve fast imaging of near space circular synthetic aperture radar (SAR) raw data with high resolution and large scene. The fast algorithm benefits from projecting the frequency domain signal after matched filtering to subimage. Quad tree structures constructed on both signal and image spaces are used to project signal recursively. Following the traversal of the quadtree, the image resolution is ameliorated gradually. The algorithm run on parallel processors and fast imaging is accomplished. Except for the ability of precise and quick imaging, large scene imaging is also achieved. The simulation results show the high quality image with good resolution.

Angular super-resolution performance is the key problem in the field of radar imaging. In this paper, we propose an approach to radar angular super-resolution through deconvolution, which is able to increase the resolution of radar image beyond the limitation of system parameters. It relies on the Bayesian formulation approach that enables to incorporate the prior information about the system and the statistical characteristics of scene. We first formulate the radar angular superresolution problem as an linear inverse problem and then convert it to a maximum a posterior (MAP) task using Bayesian theory. We then solve the MAP problem in a convex optimization framework using a shrinkage based iterative procedure, leading to algorithm that guarantees the solution to converge the global maximizer of an associated MAP criterion. Numerical experiments with synthetic data demonstrate the performance of proposed angular super-resolution algorithm.

This paper present a novel superresolution algorithm for real-beam scanning radar based on the iterative adaptive strategy. Firstly, we establish the objective function based on the minimum mean-square error (MMSE) criterion, then we build the iterative expression by update the covariance matrix. This algorithm has better superresolution performance than traditional deconvolution method. Simulation results verified the analysis before.

This paper present a superresolution algorithm for forward-looking imaging of scanning radar based on weighted least squares method. This algorithm utilized the weighted vectors to structure the objective function, and introduced the diagonal loading technique to obtain more robust superresolution result. Simulation results verified the performance of our algorithm.

Near-space Synthetic Aperture Radar (SAR) possesses useful features such as High-Resolution Wide-Swath (HRWS) and high revisiting frequency etc., which are difficult for the conventional SAR, e.g. spaceborne and airborne SAR, to simultaneously provide. Slow-speed SAR is introduced to solve the contradiction between high resolution and wide swath of Near-space SAR and a frequency domain algorithm based on Two-Step Processing Approach (TSPA) and Stolt interpolation is proposed. Simulations are provided to demonstrate the validity of the proposed algorithm.

This paper presents a two-dimensional (2-D) generalized omega-K algorithm for high-precision processing of the data of missile-borne Synthetic Aperture Radar (SAR) with constant acceleration. The almost adequate expression of the 2-D spectrum of the echo signal is obtained by the Method of Series Reversion (MSR). Then a 2-D generalized Stolt interpolation is carried out. The proposed algorithm can be applied to situations involving large dive and squint angle and high maneuvering speed. Simulations are presented to demonstrate the validity of the proposed algorithm.

This paper present an efficient matrix inverse algorithm of the recent iterative adaptive approach (IAA) in the application of real beam superresolution (RBS). Based on the inherently band structure of the covariance matrix, the computational complexity of inverse can be reduced by avoiding the computation of zero elements. To achieve this, the divide and conquer (D&C) method will be introduced to fast inverse the covariance matrix. Numerical simulations illustrate the efficiency of the proposed algorithm.

Doppler parameters, range cell migrations (RCMs), and higher order coupling terms in the raw data of translational variant bistatic synthetic aperture radar (TV-BiSAR) exhibit 2-D spatial variations. The 2-D spatial variations result in significant performance degradation of TV-BiSAR imaging. To solve these problems, an Omega- k imaging algorithm based on linearization theory is proposed in this letter. Compared with the traditional Omega- k algorithm that uses the 1-D Stolt transformation to eliminate only the spatial variations in the range direction, the proposed algorithm applies the 2-D Stolt transformations to achieve the goal of both 2-D frequency linearization and 2-D spatial-domain linearization. After the 2-D Stolt transformations, a focused image can be obtained by performing a 2-D inverse fast Fourier transform (IFFT). In the proposed Omega- k imaging algorithm, the 2-D spatial variations of the Doppler parameters, RCM, and higher order coupling terms for TV-BiSAR can be simultaneously eliminated. However, in previous publications about BiSAR imaging algorithms, the spatial variations in the azimuth direction are barely considered, which reduces the imaging accuracy. Numerical simulations verify the effectiveness of the proposed method.

Feature extraction is a key step in synthetic-aperture-radar automatic target recognition. In this letter, we propose a novel feature extraction method named sample discriminant analysis (SDA) that is based on the manifold learning theory. The method directly extracts features from 2-D image matrices rather than vectors. Furthermore, SDA preserves the neighborhood information of the original data in dimension reduction. It also makes within-class samples closer and makes between-class samples father away in a low-dimensional space. Meanwhile, a sample discriminant coefficient is employed in the method to give each sample a weight related to its location and similarity to neighboring samples. Thus, the discriminative ability of the method is improved. Experimental results based on the moving and stationary target acquisition and recognition database show that the proposed method can improve recognition performance.

With appropriate geometry configurations, bistatic synthetic aperture radar (SAR) can break through the limitations of monostatic SAR on forward-looking imaging. Thanks to such a capability, bistatic forward-looking SAR (BFSAR) has extensive potential applications, such as self-navigation and self-landing. In the mode of BFSAR with a stationary transmitter (ST-BFSAR), the two-dimensional spatial variation makes it difficult to use traditional data focusing algorithms. In this letter, an imaging algorithm based on keystone transform and nonlinear chirp scaling (NLCS) is proposed to deal with this problem. Keystone transform is used to remove the spatial variation of range cell migration. NLCS can eliminate the variation of azimuth reference function. Numerical simulations show that by combining first-order keystone transform and azimuth NLCS operation, the raw data of ST-BFSAR can be well imaged.

In translationally variant (TV) bistatic synthetic aperture radar (BSAR), 2-D spatial variation is a major problem to be tackled. In this paper, a generalized Omega-K imaging algorithm to deal with this problem is proposed. The method utilizes a point target reference spectrum of the generalized Loffeld's bistatic formula (LBF) (GLBF). Without the bistatic-deformation term, GLBF is the latest development of LBF. Similar to the monostatic case, it has a much simpler form than other point target reference spectra. Based on the spatial linearization of GLBF, the Stolt mapping relationship is derived. Different from the traditional Omega-K algorithms for monostatic SAR and translationally invariant BSAR, this approach uses a 2-D Stolt frequency transformation. Through this transformation, the method can deal with the 2-D spatial variation. It can also consider the linear spatial variation of Doppler parameters, which is usually not considered in the previous publications on bistatic Omega-K algorithms. This method can handle the cases of TV-BSAR with different trajectories, different velocities, high squint angles, and large bistatic angles. In addition, a compensation method for the phase error caused by the linearization is discussed. Numerical simulations and experimental data processing verify the effectiveness of the proposed method.

In this paper, an omega-K imaging algorithm to focus the raw data of translational invariant (TI) bistatic synthetic aperture radar (BSAR) is proposed. The method utilizes a point target reference spectrum of generalized Loffeld's bistatic formula (GLBF). Without the bistatic deformation term, GLBF is the latest development of Loffeld's bistatic formula. It is comparable in precision with the method of series reversion (MSR), but it has a much simpler form than MSR and a similar form to a monostatic case. Based on the spatial linearization of GLBF, the Stolt transformation relationship is derived. The method can consider the linear spatial variation of Doppler parameters, which is always ignored in previous publications about bistatic omega-K algorithms. This method can handle the cases of TI BSAR with high squint angles and large bistatic degrees. In addition, a compensation method for the phase error caused by the linearization is discussed. Numerical simulations and experimental data processing verify the effectiveness of the proposed method.

In this paper, a generalized frequency domain imaging algorithm is proposed for focusing data from the bistatic SAR (BSAR) with arbitrary geometry configurations. The proposed algorithm is derived from the ideal frequency-domain spectrum of arbitrary BSAR, which is achieved from the method of the series reversion (MSR). By implementing the 2-D linear regression, the process of the ideal frequency domain imaging reconstruction of arbitrary BSAR is fitted to be a two-dimensional non-uniform discrete Fourier transform (NUDFT). Then the Non-Uniform Fast Fourier Transform of type 3 (NUFFT-3) is used to compensate of the space-variance of BSAR data. Simulation results demonstrate the validity of the proposed algorithm.

Angular resolution of real aperture radar is limited by the aperture size and wavelength. In this paper, a novel angular superresolution algorithm based on Fourier-Wavelet regularized deconvolution (ForWaRD) is proposed. Orthogonal expansion and scalar shrinkage in a tandem transform domain lie at the core of the ForWaRD. Duo to deconvolution is a noise sensitive process, the method is used to deconvolve and denoise the echo signal simultaneously. We also conclude that optimum performance of this algorithm is simultaneously determined by the Fourier structure of the antenna pattern and the wavelet structure of the surface scatterers. Simulation results show that the algorithm can efficiently enhance the resolution of scanning radar under a low SNR environment.

Near space slow speed SAR owns the ability of sustainable and large-scene imaging. However, the output speed of azimuth image is slow due to the slow speed motion, which is a disadvantage for some applications. In this paper, the terrain observation by progressive scans (TOPS) mode is applied in near space slow speed SAR, and the signal properties are analyzed.

Feature extraction is a key step in synthetic aperture radar (SAR) automatic target recognition (ATR). In this paper, we propose a feature extraction algorithm based on manifold learning theory, the algorithm is named Two-dimensional Principal Component Analysis-based Two-dimensional Marginal Sample Discriminant Embedding (2DPCA-based 2DMSDE). Above all, the original SAR images are projected by 2DPCA which is effective for feature representation, the dimension of SAR images is reduced in horizontal direction and global information of the original dataset is preserved. Furthermore, 2DMSDE is employed to reduce dimension in vertical direction , preserve local information of the dataset and enhance discriminative ability. Therefore, 2DPCA-based 2DMSDE not only further compresses the dimensions of original images, but also achieves better recognition performance. Experimental results demonstrate the effectiveness of 2DPCA-based 2DMSDE.

A new ship imaging and tracking method using linear frequency modulated continuous wave (LFMCW) scanning radar in shipping lane management application is proposed. It aims at solving the problem that it is difficult for a ship to find the right shipping lane when it first enters an unknown port in bad weather conditions. A theoretical analysis is presented, demonstrating the imaging process of LFMCW radar. An image can be got after the radar scans 360°. Multi-frame images of a ship can be obtained by continuous scanning the ship area. Changes between images are used for ship tracking by multi-frame joint detection algorithm. It is then validated on an experiment. The experiment results show the efficiency of the proposed method.

Due to left/right Doppler ambiguity, and the small difference in Doppler frequencies of adjacent points in flight direction, conventional monostatic SAR can not be used for forward-looking imaging. Then bistatic SAR with transmitter and receiver mounted on different platforms is coming into the eyes of researchers. In this paper, the feasibility of forward-looking imaging using airborne bistatic SAR is studied. Firstly, the potential two-dimension resolution ability of bistatic SAR for forward-looking imaging is discussed from the aspect of iso-Doppler and iso-range line. Then the airborne bistatic SAR experiment using a side-looking airborne transmitter and a forward-looking airborne receiver is described, and the preliminary results are presented. To the authors' knowledge, this is the first bistatic forward-looking SAR experiment in the world that both transmitter and receiver are mounted on aircraft platforms.

Pulse repetition frequency (PRF) redundancy exists in near space slow speed SAR due to slow speed motion, and then the beam can point to different scene at the same position. Hence, the large scene imaging can be realized. However, the range migration and azimuth resolution is azimuth variant, which makes the imaging difficult. In this paper, imaging mode of near space slow speed SAR is presented, and the variable aperture BP imaging algorithm is proposed. Simulations are given to verify the effectiveness of the imaging scheme.

Simultaneously real-time monitoring for several distributed areas of interest is getting important in many applications. Such as security guarantee system during Olympic Games and brushfire monitoring for power grid. However, current remote sensing systems can hardly meet such steep demands. In this paper, we introduce a brand new system of Balloon SAR with Multi-spotlight illuminate mode. The new system is qualified with high resolution imaging ability for several flexible areas. Meanwhile, real time monitoring for long time duration is available here.

The generation of SAR echo is very important for both the system design and the validity test of imaging algorithm. Common time-domain-based echo generation methods are usually time-consuming and inefficient. Hence, this article will focus on the research of efficient echo generation methods. In translational variant bistatic mode, two-dimensional (2D) spatial variability is the major problem to be faced with in efficient echo generation. To solve this problem, this paper proposes an efficient echo generation method of translational variant bistatic SAR based on 2D inverse Stolt mapping. The proposed method first uses 2D FFT to generate 2D linear spectrum, and then completes nonlinearized operation of the spectrum by 2D inverse Stolt mapping. This process fully considers the 2D spatial variability in translational variant bistatic mode, which guarantees the accuracy of the generated echo. Finally, simulation results are presented to verify the validity of the proposed method.

A novel ground moving target detection and imaging model called Near-Space Slow SAR (NSS-SAR) is introduced in this paper. It is not only effective for fast-moving targets but also slow-moving and micro-moving targets, which is hardly possible for conventional airborne and spaceborne SAR. Meanwhile, this model only needs mono-channel to distinguish Doppler signature of moving targets from the competing ground clutter returns via Doppler filtering. In addition, following analysis demonstrates that the NSS-SAR also has the potential to simultaneously achieve high-resolution and wide-swath (HRWS) imaging. Simulations given at the end of this paper verify the validity of the new NSS-SAR mono-channel moving target detection and imaging model.

The angular resolution is limited by the aperture size in real beam radar. To improve the angular resolution, some deconvolution algorithms have been proposed. However, it becomes challenging to estimate the amplitude and location parameters of illuminated targets as signal-to-noise ratio decreases. Through our research, we analyze the similarities of mathematic model and physical principle between array processing and real beam imaging, then in this paper we will show how the iterative adaptive approach (IAA), a spectral estimation method, can be applied to real beam radar for superresolution. The simulation results of real beam radar will be presented to demonstrate the performance of IAA.

One-stationary bistatic side-looking synthetic aperture radar (OS-BiSAR) data are more challenging to process than the translational azimuth-invariant bistatic counterparts because 2-D spatial variances exist in OS-BiSAR. To overcome the problem, extended-keystone-transform (EKT)-nonlinear-chirp-scaling (NLCS) imaging algorithm is proposed in this letter. The key steps of the proposed algorithm are deducing the twice EKTs to deal with the 2-D spatial variance of range cell migrations and equalizing the 2-D spatial variant Doppler FM rates by using the associative NLCS. As a result, raw data of OS-BiSAR can be well focused even with the 2-D spatial variances. The simulations given at the end of this letter verify the effectiveness of this imaging algorithm.

In this paper, a frequency domain imaging algorithm is proposed for the spaceborne/airborne bistatic synthetic aperture radar (SA-BSAR) with highly squint angle. The imaging processing is carried out with the following two stages. In the first stage, the space-invariant part of the raw spectrum data is compensated by multiplying with the conjugate of the spectrum from the reference target. In the second stage, the two-dimensional space-variant component, which manifests obvious nonlinear coupling between the range frequency and the Doppler frequency in the high squint case, is effectively corrected by the two-dimensional non-uniform fast Fourier transform (NUFFT) operation. The computation burden of the proposed imaging reconstruction method is O(MN logMN), where MN is the number of the image pixels. Simulation experiments demonstrate the validity of the proposed method.

Non-uniform FFT (NUFFT) algorithm is widely applied to Communication, Medical Imaging, Radio Astronomy and so on. NUFFT contributes to SAR imaging and Near-Space SAR also becomes popular recently years. It is difficult to get accurate SAR image in Near-Space because of the obvious motion error and Wide-Swath when the vehicle is slow, so we propose an approach which uses twice NUFFT in ω - k algorithm to solve the problem in this paper. The first NUFFT can solve non-uniform sampling along the track, the second NUFFT can replace the Stolt interpolation and the final range inverse FFT in ω - k algorithm to increase the computational efficiency without reducing the accuracy of the SAR image. The new method is especially suited for Near-Space SAR imaging when the speed of the vehicle is slow.

For high-quality synthetic aperture radar (SAR) processing, Doppler centroid estimation is an essential procedure. An incorrect Doppler centroid would cause a loss of signal-to-noise ratio, an increase in the azimuth ambiguity level, and a shift in the location of the target. Based on the analysis of the range migration characteristic of the transmitter-fixed bistatic forward-looking SAR, a geometry-based Doppler centroid estimator is proposed in this letter. By searching for the range walk slope based on the minimum entropy, this method estimates the unambiguous Doppler centroid directly. Simulations validate the effectiveness of this method.

Focusing bistatic SAR data in frequency domain requires the two dimensional (2D) point target reference spectrum (PTRS). In this paper, a 2D PTRS is derived based on Lof feld's bistatic formula (LBF). The spectrum in this paper combines the influence of the Doppler centroids and frequency modulated rates of the transmitter and receiver on the total Doppler contributions. The essential of this paper is to model the Doppler contributions of the transmitter and the receiver using power series. It can be used in extreme bistatic (like hybrid spaceborne/airborne) and high squint (like bistatic forward-looking) cases. The accuracy of the PTRS is verified using numerical simulation of point target.

Doppler centroid is an important parameter for high quality synthetic aperture radar (SAR) imaging. In this paper, several geometry-based Doppler ambiguity resolvers are discussed and compared. The result show that these methods can give accurate estimation of Doppler ambiguity number, and the speed of the iterative scheme and improved Radon transform scheme is much faster than conventional Radon transform method.

Doppler rate is an important parameter in synthetic aperture radar (SAR) imaging. Incorrect Doppler rate would cause great degradation of image quality. In this paper, by estimating the Doppler rate in high-contrast scene first, and then utilizing the fact that Doppler rate is inversely proportional to range, the Doppler rate in low-contrast scene is estimated indirectly and effectively.

In this paper, a modified two-step RCMC method is proposed for SA-FBSAR. Comparing with the traditional two-step method, the sequence of the two-dimensional RCMC operations is altered to accommodate the significant nonlinearity of RCM in SA-FBSAR, and the influence of such modification on imaging process is taken into account. Simulation results with point scatterers demonstrate the validity of the proposed RCMC method.

With the ability of 3-D resolution, LASAR is more suitable for the reconnaissance and surveying applications in urban and mountain areas. However, there are some obstacles for airplane-borned LASAR. Compared with airplane, airship is more feasible for LASAR, because of its large size, long duration and huge loading capacity. Combined with MIMO and sparse array techniques, the system cost is affordable in practice. Sparsity of 3-D image should be highlighted during the course of image processing, which can reduce the algorithms' complexity and improve the image quality.

With appropriate geometry configurations, bistatic Synthetic Aperture Radar (SAR) can break through the limitations of monostatic SAR on forward-looking imaging. Thanks to such a capability, bistatic forward-looking SAR (BFSAR) has extensive potential applications. In this paper, we present a vehicle-borne BFSAR experiment, including the imaging principle, system setup, echo property and data processing result. In the experiment, the transmitter is located stationarily besides the moving receiver. The transmit and receive antennas both point to the up forward-looking area to illuminate a point target. To the authors' knowledge, this is the first result of BFSAR with stationary transmitter in the world right now.

For high-quality synthetic aperture radar (SAR) processing, Doppler centroid estimation is an essential procedure. An incorrect Doppler centroid would cause a loss of SNR, an increase in the azimuth ambiguity level, a shift in the location of the target, etc. An improved Radon-transform-based Doppler centroid estimation scheme of bistatic forward-looking SAR is proposed in this letter. First, this scheme performs edge detection on the range-compressed data and then does the coarse and fine Radon transforms to estimate the Doppler centroid. Simulations and real-data experiments validate the effectiveness of this method.

Bistatic synthetic aperture radar (BSAR) with one fixed station (OF-BSAR) can be used in wide area surveillance, interferometry and etc. This paper analyzed the spatial spectrum of OF-BSAR. Analytical expressions of the spatial spectrum was given. Using this result, we can determine the resolution performance of OF-BSAR.

This paper discusses the antenna phase center trajectory (APCT) design for the "one-active" linear-array 3D imaging SAR (LASAR). First, we discuss the principle of the one-active LASAR and demonstrate its feasibility by experiment. To describe the 3D spatial resolution of the one-active LASAR, the relationship between the 3D ambiguity function (AF) of the one-active LASAR and the system parameters is discussed in detail. Based on the analysis, we divide the APCT design into three topics: the direction of the linear array, the length of the linear array, and the switching mode of the active element [named as antenna phase center function (APCF)]. On the first topic, we conclude that, when the range, along-track, and cross-track directions are orthogonal to each other, the ambiguity region of the one-active LASAR attains minimum, and the 3D spatial resolution can be separated into the range, along-track, and cross-track resolutions. On the second topic, we find that the cross-track resolution is determined by the length of the linear array and the frequency of the carrier. To ensure that the length of the linear array is acceptable, the carrier should be W-band wave or millimeter wave. On the third topic, the effect of APCF is researched, and we find that both the periodic APCF and the pseudorandom APCF can produce 3D resolution, except for the periodic rectangle APCF. For the pseudorandom APCF and the periodic APCF with short period, the cross-range 2D AF is or can be approximated as the product of two 1D AFs in the along- and cross-track directions. Finally, the distribution of the pseudorandom APCF is optimized by the Lagrange multiplier method under the minimum variance criterion, and we find that, when the pseudorandom APCF obeys the parabolic distribution, the cross-range 2D AF is optimal.

For a high-order polynomial-phase signal (PPS), instantaneous frequency rate (IFR), which is defined as the second derivative of the phase, is estimated by using an estimator with only a second-order nonlinearity. Compared to high-order phase function (HPF), the proposed IFR estimator presents improved performance including smaller mean-squared error (MSE) and lower SNR threshold. Statistical analysis via a multivariate first-order perturbation analysis is derived for the estimate bias and MSE. Numerical results verify our analytical results.

With appropriate geometry configurations, bistatic Synthetic Aperture Radar (SAR) can break through the limitations of monostatic SAR on forward-looking imaging. With such a capability, bistatic forward-looking SAR (BFSAR) has extensive potential applications. In this paper, based on the resolution calculation using gradient theory, we give a general rule to determine the optimal geometry configuration of different modes of BFSAR. The results can be used to design BFSAR flight campaign and measure the performance of a specific BFSAR system.

This letter concerns the resolution-fusion method for linear array 3-D imaging SAR (LASAR). Limited by the length of the linear array, the cross-track resolution of LASAR is often lower than that in the along-track direction. To overcome this disadvantage, we assume that there are two LASAR systems whose trajectories are orthogonal to each other. Thus, we obtain a row low-resolution image and a column low-resolution image of the same scene. Using the orthogonal complement decomposition technique, we fuse the two images into one quasi-high-resolution image. Moreover, we find that the fusion distortion is unavoidable and contains the high-frequency component in both row and column directions. The information loss ratio is (N- 1)2/N2(Ndenotes the ratio of low resolution to high resolution). For a smooth image, the energy loss ratio is near to zero. With the increase of the noise energy, the energy loss ratio increases correspondingly. When the noise submerges the image completely, the energy loss ratio converges to the information loss ratio.

This paper discusses the antenna phase centre trajectory (APCT) design for the sparse linear array 3-D imaging SAR(LASAR). Firstly, we introduce the signal model of the sparse LASAR. Based on the model, we discuss the 3-D ambiguity function (AF) of the sparse LASAR, and reveal the relationship between the 3-D AF and the system parameters. Finally, the distribution of the pseudo-random APCF is optimized by the Lagrange multiplier method under the minimum variance criterion, and we find that when the pseudo random APCF obeys the parabolic distribution, the cross-range 2-D AF attains optimal, which has both good mainlobe and sidelobe performance.

In this paper, we present a fast 3-D imaging method for linear antenna synthetic aperture radar (LASAR). The basic idea of this technique is to consider the 3-D SAR imaging problem as tracing a surface in the observation region, since a great deal of 3-D image region contains no scatterer (such as atmosphere) or is shadowed by the other scatterers. The steps of the fast 3-D imaging method includes: Initiation, prediction, searching and recursion. Finally, some numerical experiments are presented to demonstrate the feasibility of this method. And we find that the computational cost of the fast 3-D imaging method varies according to the fluctuation of ground, and is about a dozen of times larger than that of 2-D BP algorithm.

To better understand the fundamental of the spaceborne/airborne hybrid bistatic SAR (SA-BSAR) image formation, the range cell migration (RCM) of the SA-BSAR is studied in the range-Doppler domain, where the RCM in SA-BSAR can be explicitly expressed in close form. Through the analysis of RCM relationship with target's azimuth and range position, we can find that, because of the system platforms' velocity difference along the azimuth direction, the RCM in SA-BSAR is 2D space variant. Therefore, the fundamental of frequency-domain image formation in SA-BSAR is to process the 2D space-variant RCM correction (RCMC) for nonreferent targets besides the bulk RCMC operation in the frequency domain. Furthermore, appropriate solutions in the frequency domain to remove the RCM in SA-BSAR are proposed and verified with five-point-target simulation.

This paper concerns the surface-tracing-based (STB) linear-array synthetic-aperture-radar (LASAR) 3-D imaging technique. The basic idea of this technique is to consider the 3-D SAR imaging problem as tracing a surface in the low height-resolution level. The STB 3-D imaging technique first initiates a low-resolution digital elevation map (DEM) using 3-D backprojection (BP) algorithm, predicts a higher resolution DEM from the known elevation using multivariate-interpolation technique, searches from the predicted elevation and obtains a higher resolution DEM, then repeats the prediction and searching recursively and obtains the fine-resolution DEM finally. By converting the 3-D LASAR imaging problem to a 2-D surface-tracing problem, the STB 3-D imaging technique can reduce the computational complexity by one order. The computational cost of STB 3-D imaging technique is analyzed, and we find out that the computational cost of STB 3-D LASAR imaging technique is determined by the surface prediction error and the fluctuation of ground. In particular, for normal distribution, when one of the earlier two factors is small, the computational cost is proportional to the other factor approximately. Finally, a new STB 3-D BP algorithm that implements the surface-prediction operation via multiresolution-approximation (MRA) technique (named as MRA 3-D BP algorithm) is presented. By operating the interpolation in frequency domain, the computational cost of MRA algorithm for sparse LASAR is near to that of RD algorithm for full-element LASAR.

In this paper, we discuss the mapping between the 3-D scene space and the bistatic synthetic aperture radar (SAR) image space and show that when the direction of the angular velocity of the bistatic SAR remains constant, the process of bistatic SAR imaging can be approximately modeled as a perspective operator from the 3-D scene space to the 2-D image space, and the perspective line is perpendicular to the plane determined by the composition direction of the T/R line of sight and the composition direction of the angular velocity of the T/R platform. Then, we show that the 2-D point spread function of the bistatic SAR is determined not only by the range and ldquoazimuthrdquo resolutions but also by the geometry of the bistatic SAR and the bases of the SAR image space, and the concept ldquoambiguity regionrdquo is introduced to describe the ambiguity problem in the 3-D scene space. Then, the range-Doppler algorithm is discussed, and a new translational-variant bistatic SAR imaging method is proposed, which uses the scaled inverse fast Fourier transform (IFFT) technique to eliminate the translational-variant feature of the SAR space resolution. The space truncation error of this new algorithm is discussed to analyze the depth of focus of the scaled IFFT bistatic SAR imaging algorithms, and we find that the upper bounce of the space truncation error is proportional to the square of the distance from the scatterer to the T/R platforms. Last, the effects of motion measurement errors are discussed in detail, and, through theoretical analysis and numerical experiments, we show that the absolute position measurement error, the baseline measurement error, the perpendicular (vertical) component of the absolute velocity measurement error (AVME), and the perpendicular component of the relative velocity measurement error (RVME) cause SAR image shifting in the image space mainly, and the parallel component of the AVME and the parallel component of the RVME cause the SAR image to severely defocus.

In this paper, an extended algorithm for parameter estimation of quadratic FM signal is derived by exploring the time diversity in the cubic phase (CP) function. The performance of the proposed algorithm is analyzed in terms of estimate bias and variance, and compared with other methods. Although the proposed algorithm employs a fourth-order nonlinearity which results in higher threshold SNR, it provides a number of advantages, such as low mean-square error (MSE) for the estimates at high SNR and simple extension for multicomponent signals. Extension to cubic FM signals is also discussed. The theoretical analysis is verified by the simulation results.

In this paper, the signal properties for the translational invariant case of bistatic synthetic aperture radar (SAR) based on the squint mode are deduced. Bistatic range history, point target response in time and frequency domains, Doppler properties and resolutions are presented in terms of the platform coordinates and the squint angles. The results are valid for large aperture and large squint angles. The accuracy of quadratic approximation in bistatic SAR is also discussed.

A complete vehicleborne Bistatic Synthetic Aperture Radar (SAR) imaging experiment is presented in this paper. Some new technologies were used to solve synchronization problems. An algorithm named the Glide Window Echo CFAR was used to realize time synchronization and collect the echo. High stable local oscillator was employed to achieve frequency and phase synchronization, and the PRF was adjusted adaptively in the Data Acquisition Device in the receive station. Finally, the Range-Doppler algorithm with range migration correction was using to image the certain scene successfully.

A frequency domain imaging algorithm for the hybrid spaceborne/airborne BSAR is presented. The key point of deriving the algorithm is the analytical evaluation of the system point target response's 2-D spectrum. To overcome the difficulty of resolving analytical solution for the stationary phase point, the spectrum's phase is approximated by two-order Taylor expanding around the point, which is not only in the neighborhood of the system's corresponding stationary phase point but also can be obtained analytically. Thus the approximated analytical spectrum is pretty close to the actual one. In the imaging algorithm, both range-dependent range cell migration and azimuth-dependent range cell migration are compensated in two steps: Inverse Scaled Fourier Transform which can be realized through the chirp z-transform and phase multiplication. The validity of the algorithm is demonstrated by experiment with the simulated data.

In this paper, we discuss the translational-variant feature of translational-variant bistatic SAR configuration using the space Taylor’s expansion in section II. In section III, a new translational-variant bistatic SAR imaging method is proposed, which uses scaled IFFT technique to eliminate the translational-variant feature of SAR space resolution. Finally, some numerical experiments are conducted to demonstrate the feasibility of this method and discuss the depth-of-focus of the scaled IFFT bistatic SAR imaging algorithms.

The cubic phase function (CPF) is recently proposed to estimate the instantaneous frequency rate (IFR) for the polynomial phase signals (PPS) in a Gaussian noise environment. However, for an impulse noise environment, the performance of the standard CPF degrades significantly. In addition, the resulting noise in the CPF is a mixture of the Gaussian and impulse noise even for a Gaussian input noise. Hence, a modified robust CPF algorithm based on the alpha-trimmed form of L-estimation is proposed in this paper. Extension to the robust higher-order phase function (HPF) is also derived. Simulation results demonstrate that the robust CPF outperforms the standard CPF in impulse noise and is also valid to estimate the IFR in Gaussian noise