Block sparse Bayesian learning (BSBL) has been widely used in inverse synthetic aperture radar (ISAR) imaging, which significantly improves the imaging performance by exploiting the sparse pattern information of ISAR images. However, the conventional Bayesian learning algorithm has high computational complexity, which hinders its applications to real-time processing of radar imaging. The approximate message passing (AMP) can be used to obtain a low complexity implementation of sparse Bayesian learning (SBL). However, AMP suffers from performance losses and even diverges in the case of high-Doppler resolution ISAR imaging where the measurement matrix can be highly correlated. To solve this problem, we propose a fast pattern coupled SBL ISAR imaging algorithm based on approximate message passing with unitary transformation (UTAMP). First, the estimates of the hyperparameters of sparse vector are obtained through UTAMP based SBL, and then nearest neighbor hyperparameters are coupled and updated for next iteration. With low complexity, the proposed algorithm can effectively exploit the sparse pattern information of ISAR images, and exhibits excellent convergence and imaging performance. Both simulation and real data experiments are carried out to verify the effectiveness of the proposed algorithm.

Sparse Bayesian learning (SBL) has found successful applications in interferometric inverse synthetic aperture radar (InISAR) imaging, especially in the presence of limited number of pulses or when using sparse apertures. SBL-based InISAR algorithms have been proven to be significantly superior to Fourier transform-based ones. However, the existing SBL-based algorithms are slow due to their high computational complexity. Moreover, there is also much room to improve in terms of imaging performance. In this work, leveraging the approximate message passing with unitary transformation (UAMP), we propose an InISAR imaging algorithm named UAMP-JSR (joint sparse recovery), which is much faster and delivers notably higher imaging accuracy than the existing SBL-based algorithms. Specifically, we develop a type-2 joint sparse model (JSM-2) for InISAR imaging and formulate it as a two-layer multiple measurement vectors (MMV) joint sparse problem. Based on a factor graph representation, the message passing techniques are used to efficiently solve this problem, which leads to the UAMP-JSR algorithm. Results based on extensive simulations and experiments based on the real data collected by the Pisa Radar (PIRAD) demonstrate the effectiveness and superiority of the proposed algorithm compared to existing algorithms.

This paper investigates the problem of mixed far-field (FF) and near-field (NF) source localization using a linear electromagnetic-vector-sensor array with gain/phase uncertainties. Firstly, several special fourth-order cumulant matrices are constructed, such that the shift invariance structure in the cumulant domain can be derived to estimate the DOA and polarization angles of each source at two electromagnetic vector sensors (EMVSs). Then, by computing the determinant of the coefficient matrix, the sources types can be classified with the prior knowledge of the number of both the FF and NF sources. On this basis, the range of NF sources and the DOAs of mixed sources at the phase reference point are captured subsequently. Finally, these estimates can be employed to generate the unknown gain/phase errors. Compared to the existing methods, the proposed one exploits both the spatial and polarization information of sources and provides a satisfactory parameters estimation performance under unknown phase/gain responses. Moreover, it does not need to perform any spectral search and not impose restriction on EMVSs placement, as well as realizes a more reasonable classification of the signal types. Simulations are carried out to verify the effectiveness of the proposed method.

Synthetic aperture radar (SAR) images have special physical scattering characteristics owing to their unique imaging mechanism. Traditional deep learning algorithms usually extract features from real-valued SAR images in a purely data-driven manner, which may ignore some important physical scattering characteristics and sacrifice some useful target information in SAR images. This undoubtedly limits the improvement in performance for SAR target recognition. To take full advantage of the physical information contained in SAR images, a complex-valued network guided with sub-aperture decomposition (CGS-Net) for SAR target recognition is proposed. According to the fact that different targets have different physical scattering characteristics at different angles, the sub-aperture decomposition is used to improve accuracy with a multi-task learning strategy. Specifically, the proposed method includes main and auxiliary tasks, which can improve the performance of the main task by learning and sharing useful information from the auxiliary task. Here, the main task is the target recognition task, and the auxiliary task is the target reconstruction task. In addition, a complex-valued network is used to extract the features from the original complex-valued SAR images, which effectively utilizes the amplitude and phase information in SAR images. The experimental results obtained using the MSTAR dataset illustrate that the proposed CGS-Net achieved an accuracy of 99.59% (without transfer learning or data augmentation) for the ten-classes targets, which is superior to the other popular deep learning methods. Moreover, the proposed method has a lightweight network structure, which is suitable for SAR target recognition tasks because SAR images usually lack a large number of labeled data. Here, the experimental results obtained using the small dataset further demonstrate the excellent performance of the proposed CGS-Net.

In recent years, deep learning (DL) networks have achieved considerable success in the SAR target detection task. Here, current DL-based methods, originally designed for natural images, SAR images usually have been treated as real-valued images and solely utilize the magnitude information for feature extraction. However, different from natural images, SAR images are complex-valued data. The conversion from the complex-valued images to the real-valued images potentially leads to disregarding valuable target features inherent in original SAR data. To develop the complex-valued information inherent in SAR images adequately, a target detection method for SAR images using the complex-valued network (CV-SAR-Det) is proposed in this paper. Specifically, the CV-SAR-Det mainly includes three complex-valued elements: network structure, loss function, and data augmentation. Firstly, the network adopts a fully complex-valued structure for target detection, which effectively leverages the separability characteristics of targets inherent in the complex-valued SAR image. Secondly, a complex-valued loss function (CV-Loss) is designed for the optimization of the CV-SAR-Det, which highlights the importance of the complex-valued information. Finally, to further improve the detection performance, a complex-valued data augmentation method is proposed for diversifying the training data, which also improves the robustness of the network. The advantages of the CV-SAR-Det have been demonstrated by the experimental results based on the MiniSAR-KU/KA datasets, which are superior to the latest SAR target detection methods.

A millimeter-wave radar is widely accepted by the public due to its low susceptibility to interference, such as changes in light, and the protection of personal privacy. With the development of the deep learning theory, the deep learning method has been dominant in the millimeter-wave radar field, which usually uses convolutional neural networks for feature extraction. In recent years, transformer networks have also been highly valued by researchers due to their parallel processing capabilities and long-distance dependency modeling capabilities. However, traditional convolutional neural networks (CNNs) and vision transformers each have their limitations: CNNs usually overlook the global features of images and vision transformers may neglect local image continuity, and both of them may impede gesture recognition performance. In addition, whether CNN or transformer, their implementation is hindered by the scarcity of public radar gesture datasets. To address these limitations, this paper proposes a new recognition method using a local pyramid visual transformer (LPVT) based on millimeter-wave radar. LPVT can capture global and local features in dynamic gesture spectrograms, ultimately improving the recognition ability of gestures. In this paper, we mainly carried out the following two tasks: building the corresponding datasets and executing gesture recognition. First, we constructed a gesture dataset for training. In this stage, we use a 77 GHz radar to collect the echo signals of gestures and preprocess them to build a dataset. Second, we propose the LPVT network specifically designed for gesture recognition tasks. By integrating local sensing into the globally focused transformer, we improve its capacity to capture both global and local features in dynamic gesture spectrograms. The experimental results using the dataset we constructed show that the proposed LPVT network achieved a gesture recognition accuracy of 92.2%, which exceeds the performance of other networks.