Spotlight-mode Synthetic Aperture Radar Research Articles

Residual high-order wavefront curvature error compensation for the polar format algorithm in widefield spotlight SAR imaging

ABSTRACT Polar format algorithm (PFA) is an efficient image formation algorithm for spotlight-mode synthetic aperture radar (SAR). However, the derivation of PFA relies upon the approximation of planar wavefront, and the resulting wavefront curvature error (WCE) causes severe SAR image degradation, e.g., distortion and defocus. The distortion stems from the first-order WCE and the defocus mainly comes from the quadratic WCE. To eliminate the defocusing effects, the existing space-variant post-filtering algorithms (SVPFAs) mainly compensate the quadratic term of WCE and ignore the residual high-order terms. Actually, the residual odd-order WCE will also broaden the main-lobe of the target and the residual even-order WCE will lead to the problem of sidelobe asymmetry. To tackle these issues, this letter proposes a modified SVPFA to correct both the quadratic and the residual high-order WCE. First, based on the derivation of PFA, the accurate expression of WCE is obtained. Second, the quadratic and residual high-order WCE are separated without complicated high-order two-dimensional Taylor expansion. Finally, the WCE is corrected by phase compensation and geometric distortion correction, respectively. The effectiveness of the proposed algorithm is verified by simulation experiments.

Plug-and-Play Synthetic Aperture Radar Image Formation Using Deep Priors

The reconstruction of synthetic aperture radar (SAR) images from phase history data is an ill-posed inverse problem which, in several lines of recent work, is solved by minimizing a cost function. Existing reconstruction methods use regularization to tackle the ill-posed nature of the imaging task. However, in general, these regularizers are either too simple to capture complex spatial patterns and can only promote fixed, predefined features, or lead to non-quadratic cost functions that are nontrivial to minimize. Recently emerging plug-and-play (PnP) priors technique is a flexible framework that allows forward models of imaging systems to be integrated with state-of-the-art regularizers. Inspired by this, we propose a novel PnP SAR image reconstruction framework for spotlight-mode SAR. SAR involves complex-valued reflectivities with spatial structure on the reflectivity magnitudes that can be learned and imposed as priors. This distinguishing aspect is formulated into our proposed framework. We demonstrate the use and effectiveness of a convolutional neural network (CNN) based prior model for the reconstruction of synthetic and real SAR scenes and compare the results with FFT-based, non-quadratic regularization-based, and dictionary learning-based reconstruction methods as well as a PnP framework with BM3D regularizer. Our results suggest that these deep priors enable the learning and incorporation of complicated spatial patterns more effectively than existing methods, and produce significantly improved images especially from limited observations.

Efficiency balanced matrix transpose method for sliding spotlight SAR imaging processing

Matrix transposition is a very critical operation in synthetic aperture radar (SAR) imaging systems. This study presents an improved corner turning memory solution for real-time SAR imaging processing. Based on the sub-matrix three-dimensional mapping method, the efficiency of matrix transposing is greatly increased. However, in sliding spotlight mode, a novel SAR imaging mode, the amount of raw data is increased too large to directly use the above method. In this study, a joint address mapping strategy for two pieces of double data rate 3 synchronous dynamic random-access memory is used to meet the needs of the amount of data and make up for the lack of original method. The experimental results show that this method can get an improved and balanced two-dimensional access efficiency for sliding spotlight mode SAR imaging processing.

Semiparametric Statistical Stripmap Synthetic Aperture Autofocusing

Autofocusing synthetic aperture imagery by maximizing a statistical quality metric such as contrast or sharpness is a well-documented approach in both synthetic aperture radar (SAR) literature and synthetic aperture sonar literature. It is most successfully applied in spotlight-mode SAR applications, where the assumption of spatial invariance of the corrupting phase function is strong and expressions for the gradients of various quality metrics with respect to standard error models have been calculated. Examples of application to stripmap imagery often involve sectioning images into small blocks, allowing spotlight algorithms to be patchwise applied. This paper formulates the gradient of the cost function in a manner that is consistent with the stripmap error model, inherently providing solutions that compensate for the spatial variance while simultaneously bypassing the need for subdividing an image or aperture. This paper formulates the stripmap gradient expression in conjunction with a computationally efficient imaging approach to rapidly achieve metric-maximizing solutions. Demonstrations are shown on a widebeam wideband rail-based system experiencing random jitter and on an unmanned-underwater-vehicle-mounted sonar system exhibiting artificially injected crabbing and sway errors. Results indicate that the algorithm is particularly effective at compensating for random and rapidly oscillating navigation and jitter errors, as well as sidelobes introduced by crabbing in arrayed systems.

Acceleration Model Analyses and Imaging Algorithm for Highly Squinted Airborne Spotlight-Mode SAR with Maneuvers

Synthetic aperture radar (SAR) imaging is usually performed along a straight path. In practice, variations from the ideal caused by maneuvers with acceleration are inevitable. The main problem is the complex signal processing, particularly in the highly squinted spotlight mode. In this paper, acceleration model is analyzed in more details. Impacts are obtained how the acceleration influences the Doppler parameters and space-variant terms, including the SAR resolution and azimuth bandwidth. According to the special signal properties of the acceleration case, an approximation of the range history is made to avoid using the method of series reversion (MSR) to obtain an accurate spectrum. Based on the high-accurate spectrum, a spotlight algorithm with a modified two-step approach is given. Simulation results show the effectiveness of the modified algorithm.

HF Radar Signal Processing Based on Tomographic Imaging and CS Technique

This study presents the application of a spotlight-mode synthetic aperture radar (SAR) imaging technique to the problem of high probablity target detection in high frequency (HF) radar system, attempting to improve its spatial resolution. The effects of finite aperture on resolution, sampling constraints and reconstruction over a complete angular range of 360 degrees are discussed. A Convolution Back Projection (CBP) algorithm has been applied to image reconstruction. In order to solve the range limitation of aspect angle with one radar-carrying platform, we collect data over a larger azimuthal range by making multi-aspect observations. Each straight line is a sub aperture over which we can perform the CBP algorithm. When we demand higher resolution for stationary target, it will cause blur with longer data acquisition time. Thus the application of the traditional imaging algorithm is limited. Compressed Sensing (CS) has recently attracted much interest as it can reduce the number of samples without compromising the imaging quality. Within this motivation, we discuss the applicability of CS and present the application constraint for HF radar system.

Multi-Subaperture PGA for SAR Autofocusing

For spotlight mode synthetic aperture radar (SAR) autofocusing, the traditional full-aperture phase gradient autofocus (PGA) algorithm might suffer from performance degradation in the presence of significant high-order phase error and residual range cell migration (RCM), which tend to occur when the coherent processing interval (CPI) is long. Meanwhile, PGA does not perform satisfactorily when applied directly on the stripmap data. To address these shortcomings, we present a multi-subaperture PGA algorithm, which takes advantage of the map drift (MD) technique. It smoothly incorporates the estimation of residual RCM and combines the subaperture phase error (SPE) estimated by PGA in a very precise manner. The methodology and accuracy of PGA-MD are investigated in detail. Experimental results indicate the effectiveness of PGA-MD in both the spotlight and the stripmap modes.

Predictive Quantization of Range-Focused SAR Raw Data

Synthetic aperture radar (SAR) systems create massive amounts of data which require huge resources for transmission or storage. The limited capacity of the downlink channel demands efficient onboard compression of SAR data. However, SAR raw data exhibit very little correlation which can be exploited in a compression algorithm. Range focusing is shown to increase the data correlation by exposing some of the distinctive features of the scene under surveillance. In this paper, we first present analysis of spotlight-mode SAR to show the source of the increased correlation in the range-focused data. Next, we propose two algorithms-transform-domain block predictive quantization (TD-BPQ) and transform-domain block predictive trellis-coded quantization (TD-BPTCQ)-for the compression of the range-focused data. Experimental results indicate that, at the rate of 1 bit/sample, and for similar or lower computational complexity, TD-BPQ and TD-BPTCQ outperform the best method proposed in the literature by 1.5 and 2.3 dB in signal-to-quantization-noise ratio, respectively. Similar improvements are observed for the rate of 2 bits/sample.

Space-Variant Filtering for Wavefront Curvature Correction in Polar Formatted Bistatic SAR Image

An analytical expression of the phase error from wavefront curvature in bistatic spotlight-mode synthetic aperture radar (SAR) image reconstructed via polar format algorithm (PFA) is derived in this paper. The wavefront curvature effect, which arises from the unrealistic planar wavefront assumption in bistatic PFA, induces phase error terms resulting in space-variant geometry distortion and defocusing. Based on the analytical expression of the wavefront curvature error, a space-variant filtering approach is applied to compensate for it in the polar formatted bistatic SAR image. Analysis on the scene size limit after using the correction method indicates that the derived space-variant filter is capable of extending the focused scene size of bistatic PFA considerably. Point target (PT) simulation validates the approach.

Analysis of Phase Error Effects on Stripmap SAS

It is often of interest to consider how uncompensated platform motion can degrade the ideal point scatterer response (PSR) of a synthetic aperture sonar (SAS). This information can be used to shape the design of the sonar itself as well as that of the platform carrying it. Also, knowledge of how certain types of motion affect a SAS image can reduce the time spent in troubleshooting motion estimation and compensation schemes. In the field of spotlight-mode synthetic aperture radar (SAR), the effects of phase errors across the synthetic aperture are well documented (for example, Chapter 5 of Carrara , 1995). The counterpart problem for the stripmap mode is less well developed in the literature. This paper explores the effects of uncompensated phase errors on stripmap imagery and shows that, under certain conditions, they are similar to those for spotlight mode processing.

Efficient SAR Raw Data Generation Including Low Squint Angles and Platform Instabilities

The performance of high-resolution synthetic aperture radar (SAR) systems under degrading conditions such as uncontrolled platform motion can only be analyzed and evaluated by modeling the whole imaging system through simulation. A major scientific problem, however, is the generation of spotlight mode SAR raw data because of the need for high numerical and modeling accuracy, and restricted computer time and space. Taking into account a squinted geometry and platform instabilities, the common generation of raw data in the time domain becomes time inefficient. This letter proposes a raw data generator with which the spotlight mode raw data set of an extended scene can be simulated and which combines the precision obtainable when working in the time domain with the efficiency of work in the frequency domain. The generated raw data are processed and analyzed to quantify the quality of the proposed simulation tool.

The Rolling Radar

In the search for a better rotating radar, a self-supporting array that rolls like a wheel can improve mechanical reliability and return higher-definition target resolution. The Rolling Radar concept evolved from the need to design and build a large rotating 3D radar as an alternative to multifaceted fixed radar installations, or large, expensive rotating radar systems. However, the physical act of rotating massive structures weighing many tons introduces a host of load bearing and other mechanical reliability problems. An alternative design needed to be conceptualized. The new concept places the radar array inside a large wheel attached to an axle. A smaller wheel affixed to the axle's other end provides support when both wheels simultaneously roll along electrified concentric rails in order to scan 360 degrees. The array wheel propels itself via a simple gravity drive consisting of a magnetic carriage that rides an electromagnetic rail around the inside circumference of the large array wheel. Displacement of its own weight causes the wheel (and the entire axle) to roll around the railed track with no visible external means of motion. The Rolling Radar achieves its initial objective - to improve mechanical reliability - because the array wheel is self-supporting, which eliminates the need for a large support structure and bearing. There is also no need for traditional electrical slip rings, rotating fluid couplers, motors or gearboxes. With fewer moving parts (compared with conventional rotating radars), reliability is increased and life cycle costs are reduced. Another important benefit is the Rolling Radar's electronic look back capability, and the ability to achieve higher spatial resolution than standard 3D radar. This occurs because the act of rolling about its axis and simultaneously rotating about a circumference sweeps an area larger than the array's physical aperture. The effect is similar to spotlight-mode synthetic aperture radar (SAR), a technique commonly used to produce high-resolution radar maps. The Rolling Radar concept was initially developed to address reliability, life cycle cost, and weight. Although still in the concept development phase, the Rolling Radar shows great promise of increasing functionality while improving resolution and reliability, and in its simplicity of design. Both the reliability and performance gains are in the process of being evaluated.

Motion Error Correction Scheme Using the Spatial Doppler Characteristics of the In-Scene Target in the SAR Imaging System

The success of target reconstruction in the SAR(Synthetic Aperture Radar) imaging system is greatly dependent on coherent detection. Incoherent detection appears as a multiplicative phase error to echoed signal, which consequently causes fatal degradations such as fading or dislocation of target image. In this paper, we propose a motion error correction scheme using an in-scene target to compensate for relative distance error between the radar and the target. We start by modeling from a wave equation for one point target, and then derive the complex motion error from extended overall echoed data. The proposed algorithm is also good for the correction of relatively large motion error because it can be applied repeatedly, and it is converged after each iteration. We use the spatial Doppler characteristics of the strong in-scene target to retrieve motion error, thus, we only used the partial spectral echo data corresponding to the strong in-scene target. By doing this, we can reduce computational loads and the number of iterations for the large motion error. We verify the performance of the proposed algorithm by applying it to the simulated spotlight-mode SAR data.

Terrain elevation mapping results from airborne spotlight-mode coherent cross-track SAR stereo

Coherent cross-track synthetic aperture radar (SAR) stereo is shown to produce high-resolution three-dimensional maps of the Earth surface. This mode utilizes image pairs with common synthetic apertures but different squint angles allowing automated stereo correspondence and disparity estimation using complex correlation calculations. This paper presents two Ku-band, coherent cross-track stereo collects over rolling and rugged terrain. The first collect generates a digital elevation map (DEM) with 1-m posts over rolling terrain using complex SAR imagery with spatial resolution of 0.125 m and a stereo convergence angle of 13.8/spl deg/. The second collect produces multiple DEMs with 3-m posts over rugged terrain utilizing complex SAR imagery with spatial resolutions better than 0.5 m and stereo convergence angles greater than 40/spl deg/. The resulting DEMs are compared to ground-truth DEMs and relative height root-mean-square, linear error 90-percent confidence, and maximum height error are reported.

Toward target recognition from synthetic aperture radar imagery using electromagnetics-based signatures

We develop a theory of target detection and classification from physics-based synthetic aperture radar (SAR) signatures. The target-return and clutter-return models are developed from electromagnetic theory. Both stripmap-mode and spotlight-mode SARs are treated. Adaptive-resolution processors, conventional SAR processors, and optimum likelihood-ratio target detectors are presented for multicomponent target detection, and their receiver operating characteristics are compared. Similarly, conventional and optimum likelihood-ratio processors are used for multicomponent target classification. We develop upper and lower bounds and present Monte Carlo simulations for the probabilities of correct classification.

Extended Chirp Scaling Algorithm for Spotlight SAR

Among Synthetic Aperture Radar (SAR) image formation algorithms, the chirp scaling algorithm (CSA) is computationally highly efficient and has been practically used. In this paper, how to implement the subaperture extended CSA in a high resolution spotlight mode SAR is studied in detail. The procedures such as subaperture processing and azimuth processing are discussed. According to the problems presented when using the method (A. Moreira, et al, 1996) to process spotlight SAR data, full formula expressions suitable for high squinted angles is presented. While point target simulation is analyzed, the method is validated by E-SAR raw data. The results can be directly implemented in the design of SAR processors.

Feature-enhanced synthetic aperture radar image formation based on nonquadratic regularization

We develop a method for the formation of spotlight-mode synthetic aperture radar (SAR) images with enhanced features. The approach is based on a regularized reconstruction of the scattering field which combines a tomographic model of the SAR observation process with prior information regarding the nature of the features of interest. Compared to conventional SAR techniques, the method we propose produces images with increased resolution, reduced sidelobes, reduced speckle and easier-to-segment regions. Our technique effectively deals with the complex-valued, random-phase nature of the underlying SAR reflectivities. An efficient and robust numerical solution is achieved through extensions of half-quadratic regularization methods to the complex-valued SAR problem. We demonstrate the performance of the method on synthetic and real SAR scenes.

Two-dimensional lattice linear prediction parameter estimation method and fast algorithm

This paper presents an extension of the one-dimensional (1-D) lattice (reflection coefficient) technique of linear prediction parameter estimation, first popularized by Burg, to the two-dimensional (2-D) case. The resulting fast recursive 2-D algorithm is a significant computational simplification over and an estimation improvement on previous attempts to extend the 1-D Burg linear prediction algorithm to 2-D by exploiting some newly discovered matrix structures. The technique presented here is useful for high resolution 2-D spectral analysis applications and the creation of high-resolution spotlight-mode synthetic aperture radar (SAR) imagery, as is illustrated.

Rate allocation for spotlight SAR phase history data compression

Complex phase history data in synthetic aperture radar (SAR) systems require extensive processing before useful images can be obtained. In spotlight mode SAR systems, useful images can be obtained by applying aperture weighting and inverse Fourier transform operations to SAR phase history data. In this paper, we are concerned with the compression of the complex phase history data obtained by a spotlight SAR system. We exploit knowledge of the aperture weighting function along with Fourier transform processing to attach a "gain" factor to each complex phase history data sample. This gain factor is then used to efficiently allocate bits to the phase history data during quantization. Performance evaluations are presented for this compression system relative to other existing SAR phase history data compression systems.

Synthetic-aperture-radar motion compensation and feature extraction by means of a relaxation-based algorithm

We study the problem of extracting target features by means of synthetic-aperture radar (SAR) in the presence of uncompensated aperture motion errors. A parametric data model for a spotlight-mode SAR system is established. The Cramér–Rao bounds (CRB’s) for the parameters of the data model are also derived. The CRB analysis shows that the unknown motion errors can significantly affect the accuracy of a common shift of the scatterer positions in the cross-range direction but have little effect on other target parameters, including the accuracy of the relative positions in the range direction. A relaxation-based motion compensation RELAX (MCRELAX) algorithm for estimating both target features and motion errors is devised. Simulation results show that the mean squared errors of the parameter estimates obtained by using the MCRELAX algorithm can approach the corresponding CRB’s. We also use a couple of examples to show that MCRELAX can simply be used for motion compensation only and can give good motion-compensation results.