Waveform decoupling is usually considered a more technical challenge for fully exploiting the potential benefits provided by multiple-input–multiple-output (MIMO) synthetic aperture radar (SAR) structure. Spotlighted as a promising solution to this challenge, the well-known orthogonal-waveform beamforming scheme has become increasingly popular. However, in some cases, the performance of digital beamforming (DBF) involved in this scheme may be significantly degraded due to the nonlinear array configuration under stringent space constraints. Up until now, relatively little research on robust DBF on receive in elevation has been presented for a nonlinear array configuration. To alleviate this, we here propose a least absolute shrinkage and selection operator (LASSO) regression-based DBF technique for the improved segmented phase coding (SPC) decoupling scheme. First, a generalized steering vector formation model based on a 3-D geometric vector is provided for the subsequent DBF operation. Given that, due to the nonlinear array configuration, the calculated steering vector exhibits space-varying characteristics within the azimuth pulse extension of the illumination beam, steering vector constraints for the desired signal and interferences are subsequently designed to allow for such variation within the azimuth footprint. Furthermore, the beamforming problem is addressed by generalized LASSO regression to approach the goal that providing a distortionless response and deep nulls for each desired signal and interference component within the azimuth pulse extension. Finally, we assess the feasibility and performance of the proposed LASSO regression-based DBF technique for waveform decoupling with the use of numerical simulations.

The separation of multiple transmit waveforms with time and frequency synchronization constitutes a considerable challenge for multiple-input multiple-output (MIMO) synthetic aperture radar (SAR) systems. It is well-known that aliased signal returns may be separable by digital beamforming (DBF) on receive in elevation. However, the current orthogonal-waveform beamforming schemes significantly increase the hardware complexity. Moreover, the direction of arrival (DOA) mismatch issue caused by topographical variations significantly increases the complexity of the DBF process. To alleviate these issues, we here introduce a multiple-subpulse separation and weighting synthesis (MSS-WS) echo separation framework, which is formed using segmented phase coding (SPC) waveforms. The proposed MSS-WS scheme can halve the number of interferences from far arrival angles, allowing for a reduction of the system complexity. In addition, constrained/generalized least absolute shrinkage and selection operator (LASSO) regression is exploited to form the beamformer with relatively high robustness in terms of dealing with the presence of topographical variations. The so-called LASSO-based dynamic beam response (LASSO-DBR) technique introduced here contains two parts: the source localization and the beamforming based on the designed constraint matrices. In this respect, the proposed LASSO-DBR beamformer can produce a distortionless response to the desired signal and still yield wide nulls for the unwanted interferences. Using numerical simulations, we illustrate the feasibility and performance of the proposed MSS-WS framework using the LASSO-DBR beamforming technique.

Deep learning (DL) has succeeded in the field of target detection and has been introduced into the researches of ground moving target indication (GMTI) for synthetic aperture radar (SAR) recently. Due to the lack of labeled data in SAR/GMTI, simulated data are usually employed to support the training of networks, which has proved to be a feasible way in practice. Although some simulated data are very close to the real radar data, the fact is that distribution differences between them are inevitable and always lead to a performance loss of the network. Motivated by recent advances in transfer learning, this letter proposes a new method for ground moving target detection of multichannel SAR systems, namely, subdomain adaptive residual network (SARN). It is built on the basis of ResNet18, and subdomain adaptation is introduced. During the network training, multi-kernel local maximum mean discrepancy (MK-LMMD) is minimized as well as classification error. Experiments on three-channel SAR data show that the proposed method significantly improves the detection performance as compared with CA-CFAR and the DL method.

Barrage jamming technique with controllable jamming coverage against synthetic aperture radar (SAR) systems is of great importance in electronic countermeasures. However, it is still a difficulty for the jammer to accurately impose controllable two-dimensional (2-D) local jamming on the regions of interest (ROIs). In this respect, a new parameterized and large-dynamic-range precise controllable (PLDR-PC) jamming method has been proposed in this paper to assist in solving such problems. Based on the SAR imaging properties of linear frequency modulation (LFM) case, the range and azimuth modulation factors have been well designed to generate large dynamic controllable coverage of jamming signals with high 2-D processing gain. In such a context, the PLDR-PC technique can provide the optimal power allocation and considerably reduce the jamming power while still ensuring the satisfactory performance. The proposed PLDR-PC technique can improve the jamming efficiency and considerably reduce the exposure probability of the jammer. Moreover, to improve the barrage jamming performance, the parameter estimation error model is also established to determine the simple yet valid jamming strategy in practical implementations. Finally, extensive numerical simulations in comparison with the current jamming methods have been carried out to demonstrate the effectiveness and prospect of the PLDR-PC technique against airborne/spaceborne SAR systems.

The problem of echo separation using digital beamforming (DBF) on receive for multiple-input multiple-output (MIMO) synthetic aperture radar (SAR) is of notable importance to allow for practical systems. Regrettably, current DBF-MIMO-SAR schemes, such as the short-term shift-orthogonal (STSO) scheme, are computationally cumbersome, increasing the required hardware complexity. To alleviate this problem, we here propose an improved echo separation solution for realizing a low-cost MIMO-SAR system. We detail a generic waveform design scheme as well as optimized monostatic radar waveforms (e.g., nonlinear frequency modulation (NLFM) signal) showing how these can be directly adopted in the proposed scheme to improve the imaging performance. The proposed scheme enables the number of the interference segments generated by unmatched waveforms to be halved by the use of the fast time spectrum preprocessing and segment synthesis, dramatically simplifying the array configuration and reduces the system complexity. By exploiting inter-pulse phase coding techniques, the proposed method can provide a reconfigurable waveform transmitting scheme, allowing the system resources in range frequency, elevation space, and Doppler domains to be jointly exploited for the separation of aliased signal returns. The proposed scheme is evaluated using extensive numerical and measured data sets, demonstrating the feasibility and potential of the proposed method for resource-limited spaceborne/airborne MIMO-SAR systems.

In Video synthetic aperture radar (Video SAR) system, the moving target will leave a shadow at its actual position due to Doppler effect. As the shadow of the moving target moves between Video SAR frames, the amplitudes of pixel points at the corresponding positions will jump between frames as well. According to this characteristic, a new method of Video SAR ground moving target detection and tracking based on the inter-frame amplitude temporal curves is proposed in this paper. In this method, the specially designed multiple receptive field fusion neural network model based on frame variation (MRFN-FV) is used to classify the pixel points with obvious inter-frame amplitude jumps on the whole-time axis, and then the false alarms are suppressed based on the temporal change characteristics of pixel points. Finally, the improved clustering algorithm is used to detect, locate and track the moving targets in each frame of SAR images. The effectiveness of the proposed method is verified through the measured data recorded by the THz band Video SAR system.

Ground moving target indication (GMTI), as a challenging task for synthetic aperture radar (SAR) systems, keeps drawing considerable attention. Robust principal component analysis (RPCA) aiming at separating low-rank and sparse components has been successfully employed in SAR systems for GMTI recently. However, its practical application is limited by the heavy computational burden as well as the requirement of manual parameter modification. To cope with this problem, a fast and free of presetting parameters RPCA network (RPCA-Net) is proposed for SAR-GMTI under strong clutter background. In the proposed method, a novel RPCA model is first introduced, where not only the low-rank and sparse terms but also the errors in practical SAR systems are taken into account. Moreover, the low-rank factorization plus scaled gradient descent (ScaledGD) is also employed to acquire low-rank clutter background rather than singular value decomposition (SVD). Then, we parameterize our proposed RPCA model and unfold it as a feedforward neural network (FNN) to acquire the iterative parameters through backpropagation. Compared to the GMTI methods based on traditional RPCA models, our proposed RPCA-Net can provide a higher detection ability and faster convergence without presetting parameters empirically. Experiments on two groups of measured data collected by airborne SAR systems validate the superior performance of the proposed RPCA-Net.

The echo separation issue for multiple-input multiple-output (MIMO) synthetic aperture radar (SAR) is usually regarded as a more technical challenge. Spotlighted as a promising solution to the echo separation, the well-known short-term shift-orthogonal (STSO) beamforming scheme has become increasingly popular. However, for airborne MIMO SAR systems, the digital beamforming (DBF) involved in the STSO scheme usually encounters more issues, e.g., the direction of arrival (DOA) mismatch induced by topography variation. Up to now, relatively less research on robust DBF processing has been conducted for airborne MIMO SAR systems. In this respect, an adaptive DBF technique, based on interference plus noise covariance matrix (IPNCM) reconstruction and desired signal steering vector estimation, has been proposed in this paper. IPNCM reconstruction and steering vector estimation can not only cope with the DOA mismatch problem, but also remove the desired signal component in the training data cells to increase the beamformer convergence rates. Consequently, the proposed approach really improves the array output signal-to-interference-plus-noise ratio (SINR). Moreover, numerous discussions and simulations are carried out to prove the effectiveness of proposed DBF technique under various disturbance environments. Compared with the current DBF techniques, the proposed method provides a bright application prospect for the STSO scheme.

Traditional space-time (ST) adaptive processing (STAP) theory is based on the assumption of narrowband or “zero-bandwidth,” where the decorrelation within the ST snapshot is ignored. However, with radar bandwidths increasing, this assumption becomes invalid due to the deteriorated decorrelation of the received signals within the ST snapshot. The decorrelation directly causes the dispersion of the received signals in both spatial and temporal domains, leading to the spreading of the clutter spectrum in the 2-D frequency (Doppler-spatial frequency) domain. With the spreading of the clutter spectrum, the clutter suppression notch in the traditional STAP filters is widened, resulting in a relative poor ability to detect slow-moving targets. In this article, we focus on the clutter suppression for wideband radar STAP. A generalized signal model of the ground clutter is first established for the wideband array radar. Using this outcome, we analyze the influence of bandwidth on the characteristics of the ground clutter and quantitatively describe the 2-D spreading of the ground clutter on the Doppler-spatial frequency plane. Moreover, the model of clutter covariance matrix for wideband STAP (W-STAP) is established. Finally, a 2-D keystone transform (KT) algorithm, referred to as ST KT (ST-KT), is proposed to eliminate the spreading of the ground clutter in the 2-D frequency domain caused by increasing bandwidths. Simulation results are employed to validate the theoretical analysis and verify the overperformance of the ST-KT based W-STAP method in terms of the output signal-to-clutter-plus-noise ratio (SCNR) of moving targets.

Moving target brings out the different position shifts and defocusing across the image sequences acquired by circular synthetic aperture radar (CSAR) due to the Doppler shift and range smear effects. In this paper, a novel moving target detection approach for single-channel CSAR is proposed based on deep neural network (DNN). A dual-channel densely connected convolutional network (DenseNet) in consideration of complex-valued information is exploited for distinguishing the ground clutter and moving target. In terms of limited CSAR measure data set available for training the DNN network, simulated moving target samples are generated and fused into the measured ones under the various motion parameters. Finally, experiments have demonstrated that the proposed DenseNet for single-channel CSAR system processes an accepted detection performance and effectively overcomes the insufficiency of the limited dataset applications.

The new coronavirus COVID-19 has been spreading all over the world in the last six months, and the death toll is still rising. The accurate diagnosis of COVID-19 is an emergent task as to stop the spreading of the virus. In this paper, we proposed to leverage image feature fusion for the diagnosis of COVID-19 in lung window computed tomography (CT). Initially, ResNet-18 and ResNet-50 were selected as the backbone deep networks to generate corresponding image representations from the CT images. Second, the representative information extracted from the two networks was fused by discriminant correlation analysis to obtain refined image features. Third, three randomized neural networks (RNNs): extreme learning machine, Schmidt neural network and random vector functional-link net, were trained using the refined features, and the predictions of the three RNNs were ensembled to get a more robust classification performance. Experiment results based on five-fold cross validation suggested that our method outperformed state-of-the-art algorithms in the diagnosis of COVID-19.

Due to a long coherent processing interval, moving targets are severely smeared in the UHF-band synthetic aperture radar (SAR) imagery. This further results in a low signal-to-clutter-and-noise ratio, which might lead to an unacceptable false-alarm rate in multichannel ground moving target indication. A method of moving target screening is presented in this letter, which serves to determine whether the target detected by a constant false-alarm rate detector is a real moving target. An inverse omega-K algorithm is implemented, which can recover the Doppler phase history of any isolated target within a clutter-suppressed omega-K SAR image. The recovered data are again processed into a subimage by a simple range-Doppler algorithm. Then, the subimage is refocused by azimuth autofocus processing. The sharpness of the subimage will not change after refocusing if it only contains stationary targets; otherwise, the sharpness will significantly improve. We can eliminate a false moving target by detecting this change. The proposed method is demonstrated on simulated and real multichannel UHF-band SAR data.

An efficient approach to achieve ground moving target indication (GMTI) for synthetic aperture radar (SAR) is to use the Monopulse-SAR system. This paper examines the statistics of monopulse ratio (MR) for SAR/GMTI model when complex Gaussian clutter-plus-noise is considered. The probability density function (pdf) of MR is analyzed in detail. Especially, the conditional likelihood function of MR under the null hypothesis is given in a closed-form defined by special functions. An automatic constant false-alarm rate (CFAR) detector for moving targets is provided and extended to a multi-MRD form to further improve the final detection performance. Experimental results are presented to examine the detection performance and validate the theoretical analysis.