Performance of conventional algorithms for along-track baseline multichannel synthetic aperture radar (SAR), which deal with azimuth ambiguities, ground moving target indication (GMTI) or interference, degrades severely for hybrid-baseline multichannel SAR. Space-variant interferometric phase, due to terrain elevation and hybrid-baseline, destroys the consistency among Doppler spectrum, which is the foundation of these algorithms. An image-domain signal suppression method designed for hybrid-baseline multichannel SAR is presented in this letter. With prior knowledge about digital elevation model (DEM), the proposed method recognizes desired signal on each pixel stack and suppresses others by image-domain digital beamforming (DBF), which gives out a reconstructed image. Signal classification is achieved by pixel-wise direction of arrival (DOA) estimation in image domain. Error in DEM is also considered and analyzed, which guides the signal classification. Simulation with ambiguities and two types of interference are presented, verifying the proposed method.

High-resolution wide-swath (HRWS) synthetic aperture radar (SAR) may benefit from reconstructing signals in the image domain, for taking local characteristics of the scene into account. An image-domain signal model for HRWS SAR to reconstruct nonambiguous images is presented in this article. This model makes the idea of controlling the regional reconstruction performance and taking the advantage of nonuniform scatter distribution possible. Resolution is also preserved perfectly by the proposed model despite whatever tradeoff is made between the azimuth-ambiguity-to-signal ratio (AASR) and the signal-to-noise ratio (SNR). The model is derived on backprojection (BP) images and can deal with different squint angles and beam steering strategies, thanks to the BP algorithm. Although the reconstruction is done pixel by pixel, the computational burden, which depends on beam steering and the specific constraints in reconstruction, may not severely increase. Two methods for Spotlight SAR and two methods for Stripmap SAR are also presented based on the proposed model to verify the model and demonstrate its potential. Simulations on both the point target and extended target with different squint angles are carried out to verify the proposed methods, which also verify the proposed model indirectly.

The ground-based arc synthetic aperture radar (ArcSAR) can realize panoramic observation by utilizing the circular motion of the antenna. Since the circular trajectory of the antenna brings complicated high-order range-azimuth coupling in echo signal, far-field approximation, reference range approximation, or second-order series approximation to the range model are commonly applied for the conventional fast imaging algorithms to decouple the range-azimuth coupling. However, when a wide-beam millimeter wave radar system is adopted to realize centimeter-level high-resolution imaging for near-field observation, these approximate treatments will cause severe defocusing in the image. In this paper, an improved range-Doppler (RD) imaging algorithm is proposed. The high-order Taylor series approximation to the range model is adopted to meet the error requirement in the near-field wide-beam condition. By using the series inversion method, the analytical expressions for range migration and azimuth matched filter in range-Doppler domain are derived to achieve range-azimuth decoupling and precise azimuthal focusing within the whole range swath. Simulations and real-data experiments demonstrate that the proposed algorithm can achieve fast and accurate imaging which can greatly support the millimeter wave ArcSAR in near-field observation applications.

Most of the interferometric synthetic aperture radars (InSARs) operate in zero-squint or low-squint geometry. A spatial phase ramp in focused squint SAR images leads to a higher demand for coregistration, which limits the application of InSAR. This letter investigates InSAR processing with a high squint angle. The spatial phase ramp is here directly used for the construction of the digital elevation model (DEM). The method based on back projection algorithm (BPA) is analyzed first. Then the form of the interferometric phase for BPA is derived with the existence of squint, proved to share the same form as conventional InSARs. The restriction that relates to coherence is also discussed. A processing flow without coregistration is finally presented. Simulated two-channel SAR data validates the proposed method.

Interferometric Synthetic aperture radar (InSAR) aided inertial navigation system (INS) is developed to introduce terrain elevation into registration, which gives out accumulated error of INS. Unwrapped InSAR interferogram plays a key role in previous InSAR/INS frameworks, while airborne and side-looking assumptions are usually made. However, phase unwrapping is time-consuming and high squint geometry with nonlinear trajectory is common for applications with InSAR/INS. In this article, a novel wrapped InSAR interferogram based positioning method for high-speed maneuverable platform is proposed. A novel back-projection based InSAR (BP-InSAR) signal model for slightly curved diving trajectory with high squint angle is firstly presented. Then process which generates phase gradient template from given DEM is presented, considering both the estimated and measured trajectories. Error analysis is also made to discuss the performance of the proposed method in single dual-channel observation. Finally, a sequential maximum likelihood estimator based on error analysis is implemented to deal with the remained uncertainty in single observation. Numerical experiments verify the proposed method and fully demonstrate the performance of it with curved trajectory and random errors in both position and attitude.

From the multisquint spaceborne Synthetic Aperture Radar (SAR) images, a novel method with low computing cost is proposed to perform the sea-land coarse discrimination. Both the two and three-term Low Rank and Sparse (LRS) decompositions are used, where the sparse and error terms are profited, respectively. The differences between each image pairs are then stacked and averaged. A thresholding test is finally operated. The experiment based on the real TerraSAR-X data preliminarily reveals that the land and sea areas could be roughly separated. The application includes avoiding territorial false alarms when conducting maritime surveillance in the sea-land junction region.

This paper studies positioning issue in navigation using InSAR concept. A novel method based on wrapped interferometric phase registration, which differs from common InSAR based navigation method, is proposed. This method may be useful when navigating throw hills or areas with low SAR image characteristics but rich terrain elevation features. Accurate DEM data is needed to generate ideal interferometric phase template for registration. Three dimensional positioning is achieved by estimating vertical bias and horizontal bias in order. The proposed method works well under different squint angle. Simulations are carried out to verify the method. Precision of the method under different squint angle is also exhibited in simulations.

Suppression technology of barrage jamming is an important approach to ensure the normal operation of the synthetic aperture radar (SAR) system. The detection and classification of jamming is a necessary procedure in this technology. Unsuitable thresholds set in the traditional methods may reduce the detection accuracy. In order to avoid it, this paper proposes a new method of barrage jamming detection and classification for SAR based on convolutional neural network (CNN). The signal model is constructed based on the statistical characteristics of the SAR echo signal. Based on this, a data set containing echo signals and interference signals is generated by simulation. Finally, the convolution neural network VGG16 is used to detect whether the signals in the dataset is contaminated by barrage jamming and identify the type of the interference. The experiment result illustrates that the VGG16 network trained by the frequency domain signals can effectively detect and classify the jamming signals.